Dissertations / Theses: 'Prediction of sales'

1

Lindström, Maja. "Food Industry Sales Prediction : A Big Data Analysis & Sales Forecast of Bake-off Products." Thesis, Umeå universitet, Institutionen för fysik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-184184.

Full text

Abstract:

In this thesis, the sales of bread and coffee bread at Coop Värmland AB have been studied. The aim was to find what factors that are important for the sales and then make predictions of how the sales will look like in the future to reduce waste and increase profits. Big data analysis and data exploration was used to get to know the data and find the factors that affect the sales the most. Time series forecasting and supervised machine learning models were used to predict future sales. The main focus was five different models that were compared and analysed, they were; Decision tree regression, Random forest regression, Artificial neural networks, Recurrent neural networks and a time series model called Prophet. Comparing the observed values to the predictions made by the models indicated that using a model based on the time series is to be preferred, that is, Prophet and Recurrent neural network. These two models gave the lowest errors and by that, the most accurate results. Prophet yielded mean absolute percentage errors of 8.295% for bread and 9.156% for coffee bread. The Recurrent neural network gave mean absolute percentage errors of 7.938% for bread and 13.12% for coffee bread. That is about twice as good as the models they are using today at Coop which are based on the mean value of the previous sales.
I denna avhandling har försäljningen av matbröd och fikabröd på Coop Värmland AB studerats. Målet var att hitta vilka faktorer som är viktiga för försäljningen och sedan förutsäga hur försäljningen kommer att se ut i framtiden för att minska svinn och öka vin- ster. Big data- analys och explorativ dataanalys har använts för att lära känna datat och hitta de faktorer som påverkar försäljningen mest. Tidsserieprediktion och olika mask- ininlärningsmodeller användes för att förutspå den framtida försäljningen. Huvudfokus var fem olika modeller som jämfördes och analyserades. De var Decision tree regression, Random forest regression, Artificial neural networks, Recurrent neural networks och en tidsseriemodell som kallas Prophet. Jämförelse mellan de observerade värdena och de värden som predicerats med modellerna indikerade att de modeller som är baserade på tidsserierna är att föredra, det vill säga Prophet och Recurrent neural networks. Dessa två modeller gav de lägsta felen och därmed de mest exakta resultaten. Prophet gav genomsnittliga absoluta procentuella fel på 8.295% för matbröd och 9.156% för fikabröd. Recurrent neural network gav genomsnittliga absoluta procentuella fel på 7.938% för matbröd och 13.12% för fikabröd. Det är ungefär dubbelt så korrekt som de modeller de använder idag på Coop som baseras på medelvärdet av tidigare försäljning.

APA, Harvard, Vancouver, ISO, and other styles

2

Smith, Benjamin. "Factors affecting the annual unit sales volume of combines in the United States." Thesis, Kansas State University, 2012. http://hdl.handle.net/2097/35264.

Full text

Abstract:

Master of Agribusiness
Department of Agricultural Economics
Allen M. Featherstone
In the United States, accurately predicting the agricultural industry’s future demand for new farm machinery is a complicated, challenging and ever-changing issue. To compound the matter; as the size of large farm machinery continues to increase, the annualized sales volume is decreasing over time. This thesis also finds that recent mandates applicable to the Environmental Protection Agency (EPA) diesel engine emission compliance and the Internal Revenue Service (IRS) Section 179 tax code may help with forecasting the demand for farm machinery on an annual basis. This thesis evaluates factors that affect the annual unit demand of combines in the United States. Due to the lack of published literature on this specific topic, a survey of John Deere dealership sales professionals who have had recent experience selling new combines to farmers was used. This perspective brings to light factors that impact industry demand for new combines. This study results in an empirical regression model with independent variables based on the survey results. A thorough understanding of the independent variables can aid in predicting the future demand for combines. This work indicates that forty years of historical data proves to provide enough variability such that statistically significant variables are identified to accurately predict future sales. Statistically significant factors that affect the annual unit sales volume of combines in the United States include: Interest Rate, Net Cash Income, IRS Section 179 Tax Code, Planted Acres and Combine Capacity. Future industry demand is predicted by applying forecasted estimates to the model’s applicable independent variables.

APA, Harvard, Vancouver, ISO, and other styles

3

Jones, Peter Charles. "The development of a model and instrument for the measurement of personality and prediction of performance sales roles." Thesis, University of Wolverhampton, 2000. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.310708.

Full text

Abstract:

The study examines the relationship between personality and work performance in sales roles. Beginning with an examination of the major trait personality models and instruments it examines the reasons that personality questionnaires have had relatively little success in predicting work performance. The review concludes that instruments based on the behaviours which distinguish between the most effective and less effective sales performers are likely to be better predictors of work behaviour than traditional measures of a broader model of personality. A detailed work analysis was conducted based with sales staff, their managers, directors and customers. To this was added documentary evidence from job advertisements, training manuals and a sales diary. Following open coding the 3565 behaviours which emerged were subjected to a process of reduction to produce two pilot behavioural questionnaires with a total of 717 questionnaire items. Detailed information about each item was gathered, including the relationship between item response and work performance, the capacity to discriminate sales staff from other groups and the relative responses of gender, age and ethnicity groups. A second pilot was conducted with studies of concurrent and construct validity being carried out and the data was subjected to oblique factor analysis and cluster analysis techniques. A final instrument of 153 items was prepared, measuring a new hierarchical model of personality in sales roles comprised of 8 lower order factors, three higher order factors and a general factor. The new instrument also measured attempts to distort the profiles given. The new instrument was completed by sales staff from a number of sales roles and measures of their work performance were also taken. Construct validity studies were undertaken with Eysenck's Personality Questionnaire, The Manchester Personality Questionnaire and the NEO-PI questionnaire. In addition the instrument was completed by two samples of none-sales staff and measures of their work performance and career advancement were taken. It was found that in general the work behaviour based instrument was no more successful than broader personality questionnaire at predicting work success. The possible reasons for these findings, for example criterion contamination and an incomplete sampling of behaviours or routes are discussed.The performance of a number of groups on the instrument are also reported on the basis of gender, ethnicity, age and first language.

APA, Harvard, Vancouver, ISO, and other styles

4

Boulin, Juan Manuel. "Call center demand forecasting : improving sales calls prediction accuracy through the combination of statistical methods and judgmental forecast." Thesis, Massachusetts Institute of Technology, 2010. http://hdl.handle.net/1721.1/59159.

Full text

Abstract:

Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management; and, (S.M.)--Massachusetts Institute of Technology, Engineering Systems Division; in conjunction with the Leaders for Global Operations Program at MIT, 2010.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 79-81).
Call centers are important for developing and maintaining healthy relationships with customers. At Dell, call centers are also at the core of the company's renowned direct model. For sales call centers in particular, the impact of proper operations is reflected not only in long-term relationships with customers, but directly on sales and revenue. Adequate staffing and proper scheduling are key factors for providing an acceptable service level to customers. In order to staff call centers appropriately to satisfy demand while minimizing operating expenses, an accurate forecast of this demand (sales calls) is required. During fiscal year 2009, inaccuracies in consumer sales call volume forecasts translated into approximately $1.1M in unnecessary overtime expenses and $34.5M in lost revenue for Dell. This work evaluates different forecasting techniques and proposes a comprehensive model to predict sales call volume based on the combination of ARIMA models and judgmental forecasting. The proposed methodology improves the accuracy of weekly forecasted call volume from 23% to 46% and of daily volume from 27% to 41%. Further improvements are easily achievable through the adjustment and projection processes introduced herein that rely on contextual information and the expertise of the forecasting team.
by Juan Manuel Boulin.
S.M.
M.B.A.

APA, Harvard, Vancouver, ISO, and other styles

5

Jesperson, Sara. "Defining and predicting fast-selling clothing options." Thesis, Linköpings universitet, Statistik och maskininlärning, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-158190.

Full text

Abstract:

This thesis aims to find a definition of fast-selling clothing options and to find a way to predict them using only a few weeks of sale data as input. The data used for this project contain daily sales and intake quantity for seasonal options, with sale start 2016-2018, provided by the department store chain Åhléns. A definition is found to describe fast-selling clothing options as those having sold a certain percentage of their intake after a fixed number of days. An alternative definition based on cluster affiliation is proven less effective. Two predictive models are tested, the first one being a probabilistic classifier and the second one being a k-nearest neighbor classifier, using the Euclidean distance. The probabilistic model is divided into three steps: transformation, clustering, and classification. The time series are transformed with B-splines to reduce dimensionality, where each time series is represented by a vector with its length and B-spline coefficients. As a tool to improve the quality of the predictions, the B-spline vectors are clustered with a Gaussian mixture model where every cluster is assigned one of the two labels fast-selling or ordinary, thus dividing the clusters into disjoint sets: one containing fast-selling clusters and the other containing ordinary clusters. Lastly, the time series to be predicted are assumed to be Laplace distributed around a B-spline and using the probability distributions provided by the clustering, the posterior probability for each class is used to classify the new observations. In the transformation step, the number of knots for the B-splines are evaluated with cross-validation and the Gaussian mixture models, from the clustering step, are evaluated with the Bayesian information criterion, BIC. The predictive performance of both classifiers is evaluated with accuracy, precision, and recall. The probabilistic model outperforms the k-nearest neighbor model with considerably higher values of accuracy, precision, and recall. The performance of each model is improved by using more data to make the predictions, most prominently with the probabilistic model.

APA, Harvard, Vancouver, ISO, and other styles

6

Fišer, Karel. "Ocenění společnosti LINET spol. s.r.o." Master's thesis, Vysoká škola ekonomická v Praze, 2012. http://www.nusl.cz/ntk/nusl-150086.

Full text

Abstract:

This Master thesis deals with the valuation of the company Linet spol. s.r.o. to the date December 8, 2012. As a valuation method was used the earnings valuation method based on discounted free cash flow to the company. Except of value of the company, there is determined internal value of private equity. Financial analysis of the company and strategic analysis are applied in the process of valuation. Process of valuation is based on financial plan and another input variables such as avarage cost of capital. Delivery of the statement of the company value is followed by valuation. A single-factor sensitivity analysis has been carried out too. It focuses on individual factors that affect the final value of the company.

APA, Harvard, Vancouver, ISO, and other styles

7

Boyapati, Sai Nikhil, and Ramesh Mummidi. "Predicting sales using Machine Learning Techniques." Thesis, Blekinge Tekniska Högskola, Institutionen för datavetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-20237.

Full text

APA, Harvard, Vancouver, ISO, and other styles

8

Voráček, Lukáš. "Quantitative approach to short-term financial planning." Master's thesis, Vysoká škola ekonomická v Praze, 2011. http://www.nusl.cz/ntk/nusl-113571.

Full text

Abstract:

The aim of this study is to certify the legitimacy of employing quantitative methods in the day-to-day business practice. The task is approached as a case study of a real-life financial planning process. I work with the financial data of POS Media Czech Republic (a media company providing point-of-sale advertising solutions). My intention is to simulate the projection of a pro forma income statement with the use of quantitative methods. More specifically, I am applying time series prediction techniques in order to forecast POS Media's sales. The goal is, first, to demonstrate that quantitative techniques can be handled even with limited statistical background and, second, to discuss the relevancy of the obtained results. In the methodical part of my thesis I deal with the theoretical aspects of financial planning. I further describe various methods of sales forecasting (qualitative vs. quantitative). Special emphasis is put on time series prediction methods. In the application part I provide a short description of POS Media and its business. I use time series decomposition techniques to predict POS Media's sales in 2012. Consequently, I outline the rest of the pro forma income statement.

APA, Harvard, Vancouver, ISO, and other styles

9

Siwerz, Robert, and Christopher Dahlén. "Predicting sales in a foodstore department using machine learning." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-208888.

Full text

Abstract:

Prediction of sales is an important field in the food industry and dueto new technologies it has recently gained alot of attention in order toimprove business operations and profitability. However, historicallythe industry has relied on traditional statistical models but in recentyears more advanced machine learning methods has gained traction.This study aims to compare three machine learning methods forsales prediction in the food industry: Multilayer Perceptron (MLP),Support Vector Machine (SVM) and Radial Basis Function Network(RBFN). The methods were compared in terms of their prediction accu-racy on daily sales in a food store department. The performance of themodels was determined using the performance measures: Mean Aver-age Percentage Error (MAPE) and Root Mean Squared Error (RMSE).The results show that the SVM performed lower error measuresthan the other two methods. The repeated measure analysis of vari-ance (rANOVA) was used in order to determine if there was a differ-ence between the methods. The test indicated a statistical significantdifference between the afformentioned methods.
Försäljningsprediktion är ett viktigt område inom livsmedelsindustrin och tack vare nya teknologier har området nyligen fått stor uppmärksamhet i syfte att förbättra affärsverksamheten och lönsamheten i en mataffär. Historiskt sett har livsmedelsindustrin dock använt sig av traditionella statistiska modeller men under senare år har mer avancerade maskininlärningsmetoder vunnit mark. Denna studie ämnar att jämföra tre maskininlärningsmetoder för försäljningsprognostisering inom livsmedelsindustrin: Multilayer Perceptron med hjälp av backpropagation (MLP), Support Vector Machine (SVM) och Radial Basis Function Network. Metoderna jämfördes med avseende på deras prediktionsträffsäkerhet på daglig försäljning i en butiksavdelning och mättes med hjälp av mätningsvärktygen: medelprocentfelet (MAPE) och rotmedelfelet (RMSE). Resultaten visar att SVM presterade lägre prediktionsfel än de andra två metoderna. Upprepad variansanalys (rANOVA) användes för att avgöra om det förelåg någon skillnad mellan metoderna. Testet indikerade en statistiskt signifikant skillnad mellan metoderna.

APA, Harvard, Vancouver, ISO, and other styles

10

Aronsson, Henrik. "Modeling strategies using predictive analytics : Forecasting future sales and churn management." Thesis, KTH, Skolan för informations- och kommunikationsteknik (ICT), 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-167130.

Full text

Abstract:

This project was carried out for a company named Attollo, a consulting firm specialized in Business Intelligence and Corporate Performance Management. The project aims to explore a new area for Attollo, predictive analytics, which is then applied to Klarna, a client of Attollo. Attollo has a partnership with IBM, which sells services for predictive analytics. The tool that this project is carried out with, is a software from IBM: SPSS Modeler. Five different examples are given of what and how the predictive work that was carried out at Klarna consisted of. From these examples, the different predictive models' functionality are described. The result of this project demonstrates, by using predictive analytics, how predictive models can be created. The conclusion is that predictive analytics enables companies to understand their customers better and hence make better decisions.
Detta projekt har utforts tillsammans med ett foretag som heter Attollo, en konsultfirma som ar specialiserade inom Business Intelligence & Coporate Performance Management. Projektet grundar sig pa att Attollo ville utforska ett nytt omrade, prediktiv analys, som sedan applicerades pa Klarna, en kund till Attollo. Attollo har ett partnerskap med IBM, som saljer tjanster for prediktiv analys. Verktyget som detta projekt utforts med, ar en mjukvara fran IBM: SPSS Modeler. Fem olika exempel beskriver det prediktiva arbetet som utfordes vid Klarna. Fran dessa exempel beskrivs ocksa de olika prediktiva modellernas funktionalitet. Resultatet av detta projekt visar hur man genom prediktiv analys kan skapa prediktiva modeller. Slutsatsen ar att prediktiv analys ger foretag storre mojlighet att forsta sina kunder och darav kunna gora battre beslut.

APA, Harvard, Vancouver, ISO, and other styles

11

Chirico, Kristina Eva Lewis Philip M. "Predicting objective measures of performance." Auburn, Ala., 2005. http://hdl.handle.net/10415/1286.

Full text

APA, Harvard, Vancouver, ISO, and other styles

12

Hejczyk, Katarzyna Ewa. "Application of crystal structure prediction to salts and cocrystals." Thesis, University of Cambridge, 2010. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.608946.

Full text

APA, Harvard, Vancouver, ISO, and other styles

13

Mohamed, S. "Computational crystal structure prediction and experimental characterisation of organic salts." Thesis, University College London (University of London), 2011. http://discovery.ucl.ac.uk/1318096/.

Full text

Abstract:

Approximately half of all pharmaceutical drugs are marketed as salts. This thesis pioneers the application of computational crystal structure prediction to organic salts containing the commonly used chloride or carboxylate counterions, and assesses the extent to which the theoretical calculations can be used to aid experimental efforts targeting organic salts. A screen for multi-component solid forms of pyridine and 4- dimethylaminopyridine (DMAP) using a range of dicarboxylic acids led to one novel cocrystal of pyridine and six novel salts of DMAP. All novel crystal structures were solved using single crystal X-ray diffraction. At a simplistic level, salts differ from cocrystals in the position of the acidic proton within the crystal. For a selected set of structures, periodic ab initio calculations were shown to be useful in suggesting the observed N+-H (salt) or O-H (cocrystal) covalent bond as the preferred atomic connectivity. Modelling the same crystal structures by lattice energy minimisation using a distributed multipole based electrostatic model proved successful if the correct proton connectivity was used. The observed structures of a model salt, cocrystal and disordered salt-cocrystal system were found to be the most stable or almost equienergetic with the most stable structure in the predicted crystal energy landscapes. When the predictions were repeated using molecular structures with the wrong proton connectivity, the energetic ranking of the structure got worse. The computational model of crystal structure prediction was successfully used to rationalise the different polymorphic and hydration behaviour of the pharmacologically active amantadine hydrochloride and memantine hydrochloride salts. Finally, a similar methodology was applied to 1,8-naphthyridinium fumarate and the calculations performed as part of the fifth international blind test of crystal structure prediction. Overall, the success in modelling the crystal structures of carboxylate and chloride salts illustrates the promise of crystal structure prediction in aiding experimental efforts of organic salt selection.

APA, Harvard, Vancouver, ISO, and other styles

14

Culbertson, Jason D. "PREDICTING SALES PERFORMANCE: CONSIDERING NONLINEAR RELATIONSHIPS BETWEEN GMA, PERFORMANCE, AND EFFECTIVENESS." Wright State University / OhioLINK, 2011. http://rave.ohiolink.edu/etdc/view?acc_num=wright1308613576.

Full text

APA, Harvard, Vancouver, ISO, and other styles

15

Lam, Stephen Tsz Tang. "Accelerated atomistic prediction of structure, dynamics and material properties in molten salts." Thesis, Massachusetts Institute of Technology, 2020. https://hdl.handle.net/1721.1/129108.

Full text

Abstract:

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Nuclear Science and Engineering, September, 2020
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 122-142).
Various advanced nuclear reactors including fluoride high-temperature salt-cooled reactors (FHRs), molten salt reactors (MSRs) and fusion devices have proposed to use molten salt coolants. However, there remain many uncertainties in the chemistry, dynamics and physicochemical properties of many salts, especially over the course of reactor operation, where impurities are introduced, and compositional and thermodynamic changes occur. Density functional theory (DFT) and ab initio molecular dynamics (AIMD) were used for property, structure and chemistry predictions for a variety of salts including LiF, KF, NaF, BeF2, LiCl, KCl, NaCl, prototypical Flibe (66.6%LiF-33.3%BeF2), and Flinak (46.5%LiF-11.5NaF-42%KF). Predictions include thermophysical and transport properties such as bulk density, thermal expansion coefficient, bulk modulus, and diffusivity, which were compared to available experimental data.
DFT consistently overpredicted bulk density by about 7%, while all other properties generally agreed with experiments within experimental and numerical uncertainties. Local structure was found to be well predicted where pair distribution functions showed similar first peak distances (+ 0.1 A) and first shell coordination numbers (+ 0.4 on average), indicating accurate simulation of chemical structures and atomic distances. Diffusivity was also generally well predicted within experimental uncertainty (+20%). Validated DFT and AIMD methods were applied to study tritium in prototypical salts since it is an important corrosive and diffusive impurity found in salt reactors. It was found that tritium species diffusivity depended on its speciation (TF vs. T2), which was related to chemical structures formed in Flibe and Flinak salts. Further, predictions allowed comparison with and interpretation of past contradictory experimental results found in the literature.
Lastly, robust neural network interatomic potentials (NNIPs) were developed for LiF and Flibe. The LiF NNIP accurately reproduced DFT calculations for pair interactions, solid LiF and liquid molten salt. The Flibe NNIP was developed for molten salt at the reactor operating temperature of 973K and was found to reproduce local structures calculated from DFT and showed good stability and accuracy during extended MD simulation. Ab initio methods and NNIPs can play a major role in advanced reactor development. Combined with experiments, these methods can greatly improve fundamental understanding and accelerate materials discovery, design and selection.
by Stephen Tsz Tang Lam.
Ph. D.
Ph.D. Massachusetts Institute of Technology, Department of Nuclear Science and Engineering

APA, Harvard, Vancouver, ISO, and other styles

16

Gogus, Cigdem. "Understanding and predicting consumers' participation in mobile sales promotions : an extended reasoned action approach." Thesis, University of Reading, 2009. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.528834.

Full text

Abstract:

The advent of the mobile technologies is transforming the relationships between the firm and its customers. The mobile phone, which has become indispensable to many people, is increasingly being used as a unique and innovative marketing channel that enables personal, interactive, anytime, anywhere communication. Although there have been some studies in mobile marketing, most of them concentrate on consumers‟ perceptions of and attitudes towards mobile marketing. There is a scarcity of research that examines consumers behaviour in relation to mobile marketing communications. This thesis develops and tests a model, based on an extended version of Fishbein and Ajzen's (1975) theory of reasoned action, to understand and predict consumers participation in mobile sales promotions. Drawing from literature in the fields of sales promotion, mobile marketing and prediction of behaviour, the model proposes four main influences on consumers participation behaviour: attitude, subjective norm, behavioural intention, and past behaviour. The results of the empirical study, a two-stage large-scale survey conducted with the participants of major mobile sales promotion campaign, show support for the effects of attitude, behavioural intention and past behaviour on consumers‟ participation in mobile sales promotions. The study reported in this thesis makes a number of contributions to marketing research; specifically to the understanding of consumer response to mobile sales promotions. From a theoretical perspective, the extensions made to the theory of reasoned action (a multidimensional attitudinal structure instead of a uni-dimensional one; a positive influence from attitudes to subjective norms instead of independency between these constructs; and the inclusion of the past behaviour construct) support earlier empirical findings that the structure underlying Fishbein and Ajzen‟s theory is richer in content and more complex than it is originally stated. From a managerial perspective the findings show that attitude, behavioural intention, and past behaviour are the three most important factors in determining consumers‟ participation in mobile sales promotions. From a methodological perspective, the use of Partial Least Squares (PLS), a robust structural equation modelling technique mainly concerned with “prediction”, is found to be useful in understanding the antecedents of behaviour under inquiry as well as predicting it; a finding that contributes to a greater understanding of the relevance of PLS in marketing based business studies.

APA, Harvard, Vancouver, ISO, and other styles

17

Dean, Suzanne Lee. "Heterogeneous versus Homogeneous Measures:A Meta-Analysis of Predictive Efficacy." Wright State University / OhioLINK, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=wright1452866556.

Full text

APA, Harvard, Vancouver, ISO, and other styles

18

Yassin, Adel Taha. "The Vertical Distribution of Salts in a Soil Profile During the Drainage Process." DigitalCommons@USU, 1986. https://digitalcommons.usu.edu/etd/4642.

Full text

Abstract:

The purpose of this study was to develop a model to predict water extraction patterns and therefore salt distribution patterns in a one dimensional homogeneous soil profile for a specified root distribution . Water extraction was simulated as a function of the total potential and the root density at any level of the profile. Salt redistribution caused by irrigation was simulated by assuming a partial and proportional displacement of the water in each soil layer. A computer program was written for the model in Fortran language and implemented on the Vax. To evaluate the performance of the model, test studies were carried out in the laboratory using two lysimeters and wheat as a crop. A neutron probe and the four-probe electrode method were used to follow the change in the soil moisture and the salinity in the profile during the growing season. Comparisons were made between the measured and simulated values of water content and salinity. Application of the model results and recommendations for further research were suggested to improve the performance of the model.

APA, Harvard, Vancouver, ISO, and other styles

19

Amethier, Patrik, and André Gerbaulet. "Sales Volume Forecasting of Ericsson Radio Units - A Statistical Learning Approach." Thesis, KTH, Matematisk statistik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-288504.

Full text

Abstract:

Demand forecasting is a well-established internal process at Ericsson, where employees from various departments within the company collaborate in order to predict future sales volumes of specific products over horizons ranging from months to a few years. This study aims to evaluate current predictions regarding radio unit products of Ericsson, draw insights from historical volume data, and finally develop a novel, statistical prediction approach. Specifically, a two-part statistical model with a decision tree followed by a neural network is trained on previous sales data of radio units, and then evaluated (also on historical data) regarding predictive accuracy. To test the hypothesis that mid-range volume predictions of a 1-3 year horizon made by data-driven statistical models can be more accurate, the two-part model makes predictions per individual radio unit product based on several predictive attributes, mainly historical volume data and information relating to geography, country and customer trends. The majority of wMAPEs per product from the predictive model were shown to be less than 5% for the three different prediction horizons, which can be compared to global wMAPEs from Ericsson's existing long range forecast process of 9% for 1 year, 13% for 2 years and 22% for 3 years. These results suggest the strength of the data-driven predictive model. However, care must be taken when comparing the two error measures and one must take into account the large variances of wMAPEs from the predictive model.
Ericsson har en väletablerad intern process för prognostisering av försäljningsvolymer, där produktnära samt kundnära roller samarbetar med inköpsorganisationen för att säkra noggranna uppskattningar angående framtidens efterfrågan. Syftet med denna studie är att evaluera tidigare prognoser, och sedan utveckla en ny prediktiv, statistisk modell som prognostiserar baserad på historisk data. Studien fokuserar på produktkategorin radio, och utvecklar en två-stegsmodell bestående av en trädmodell och ett neuralt nätverk. För att testa hypotesen att en 1-3 års prognos för en produkt kan göras mer noggran med en datadriven modell, tränas modellen på attribut kopplat till produkten, till exempel historiska volymer för produkten, och volymtrender inom produktens marknadsområden och kundgrupper. Detta resulterade i flera prognoser på olika tidshorisonter, nämligen 1-12 månader, 13-24 månader samt 25-36 månder. Majoriteten av wMAPE-felen för dess prognoser visades ligga under 5%, vilket kan jämföras med wMAPE på 9% för Ericssons befintliga 1-årsprognoser, 13% för 2-årsprognerna samt 22% för 3-årsprognoserna. Detta pekar på att datadrivna, statistiska metoder kan användas för att producera gedigna prognoser för framtida försäljningsvolymer, men hänsyn bör tas till jämförelsen mellan de kvalitativa uppskattningarna och de statistiska prognoserna, samt de höga varianserna i felen.

APA, Harvard, Vancouver, ISO, and other styles

20

Kane-Sellers, Marjorie Laura. "Predictive models of employee voluntary turnover in a North American professional sales force using data-mining analysis." [College Station, Tex. : Texas A&M University, 2007. http://hdl.handle.net/1969.1/ETD-TAMU-1486.

Full text

APA, Harvard, Vancouver, ISO, and other styles

21

Moraes, Renan Manhabosco. "Aplicações de técnicas multivariadas na área comercial de uma empresa de comunicação." reponame:Biblioteca Digital de Teses e Dissertações da UFRGS, 2017. http://hdl.handle.net/10183/173130.

Full text

Abstract:

A mudança de comportamento dos consumidores através do advento da tecnologia e das redes sociais gera um grande empoderamento dos mesmos, alterando substancialmente a forma de relacionamento das empresas com seu público final. Atentas a este mercado, as empresas de mídia passam por profundas mudanças, tanto do ponto de vista da entrega de conteúdo ao seu público, quanto no seu formato administrativo, estratégico e financeiro. Sendo assim, a presente dissertação apresenta abordagens apoiadas em técnicas multivariadas para composição de equipes comerciais e de remuneração dos times de venda de uma empresa de comunicação. No artigo 1, objetiva-se gerar um modelo para estimar a premiação comercial das equipes de venda das rádios do Grupo RBS. Para tanto, inicialmente geram-se agrupamentos das emissoras de rádio do Grupo RBS no estado do Rio Grande do Sul e de Santa Catarina com base nos seus perfis de similaridades. Para cada cluster gerado, gera-se uma regressão linear múltipla da premiação comercial validado através de validação cruzada por intermédio do R2 ajustado e Mean Absolute Percentage Error (MAPE). O segundo artigo aborda a clusterização dos top clientes do Grupo RBS e o impacto na composição das equipes comerciais por meio do método da seleção de variáveis. As 7 variáveis originais foram avaliadas através do método de seleção de variáveis “Omita uma variável por vez”; o melhor Silhouette Index (SI) médio, métrica utilizada para avaliar a qualidade dos agrupamentos gerados, foi obtido quando 3 variáveis foram retidas. Os agrupamentos gerados por tais variáveis refletem o comportamento de compra de mídia dos clientes; os agrupamentos foram considerados satisfatórios quando avaliados por especialistas do Grupo RBS.
The change in the behavior of consumers with the advent of technology and social networks generates a great empowerment of themselves, substantially altering the relationship form of companies to their final audience. Attentive to this market, media companies undergo profound changes, both from the point of view of delivering content to their audience, as well as in their administrative, strategic and financial format. Thus, the present dissertation presents approaches supported by multivariate techniques for the composition of commercial and remuneration teams of the sales group of a communication company. In article 1, the objective is to generate a model to estimate the commercial awards of the sales teams of the RBS Group radios. To do this, we initially generate groupings of radio stations from the RBS Group in the state of Rio Grande do Sul and Santa Catarina based on their profiles of similarities. For each cluster generated, a multiple linear regression of the commercial award is generated, validated through cross validation through the adjusted R2 and Mean Absolute Percentage Error (MAPE). The second article addresses the clustering of RBS Group top clients and the impact on the composition of business teams through the variable selection method. The original 7 variables were evaluated through the variable selection method "Omit one variable at a time"; the best Silhouette Index (SI) average, metric used to evaluate the quality of the generated clusters, was obtained when 3 variables were retained. Clusters generated by such variables reflect customers' buying behavior of media; the clusters were considered satisfactory when evaluated by RBS Group experts.

APA, Harvard, Vancouver, ISO, and other styles

22

Hesqua, Rene. "Incremental validity of assessment centre exercise ratings over and above general mental ability and personality traits in predicting financial intermediaries regulatory examination success and sales performance." Master's thesis, University of Cape Town, 2018. http://hdl.handle.net/11427/29808.

Full text

Abstract:

The present study explored the optimal selection of financial intermediaries in the South African insurance services industry. We examined the potential predictive value of competency-based selection assessment centre (AC) exercises, when used in combination with other traditional measures (e.g., interviews, work style interest questionnaire, general mental ability and personality traits) in an incremental validity study. Moreover, the study expanded the criterion domain by considering prediction models for multiple outcome measures, including examination success of intermediaries and their objective sales performance. The sample consisted of 425 intermediaries in the middle and affluent market segments in South Africa. We reduced the large number of potential predictors through principle component analysis and subsequently conducted hierarchical regression analysis. Results showed that when the independent variables were restricted to correlating predictors as part of the further analysis, assessment centre exercises, especially the role-play, had a significant incremental effect (ΔR²=.07, ��<.05) over general mental ability (GMA) in predicting examination marks. Personality traits (e.g., emotional stability) and assessment centre ratings contributed significantly to examination success. To predict sales performance, the personality measure incremented GMA and AC exercises (ΔR²=.08, p < .05). GMA and AC exercises did not contribute in predicting sales performance and could not add incremental validity (ΔR² = .01, p > .05). The work style questionnaire (ΔR² = .03, p < .05) and interview (ΔR² = .012, p < .05) selection measures incremented both AC exercises and personality traits in predicting sales performance. The study contributes to our understanding of predictor combinations when academic and objective performance criteria are considered in a specific applied setting (e.g., a niche industry). The practical implications of the findings are that the validity of the assessment centre exercises together with other measures could be enhanced since it can predict different aspects of performance.

APA, Harvard, Vancouver, ISO, and other styles

23

Rangavajjula, Santosh Bharadwaj. "Design of information tree for support related queries: Axis Communications AB : An exploratory research study in debug suggestions with machine learning at Axis Communications, Lund." Thesis, Blekinge Tekniska Högskola, Institutionen för datalogi och datorsystemteknik, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-16826.

Full text

Abstract:

Context: In today's world, we have access to so much data than at any time in the past with more and more data coming from smartphones, sensors networks, and business processes. But, most of this data is meaningless, if it's not properly formatted and utilized. Traditionally, in service support teams, issues raised by customers are processed locally, made reports and sent over in the support line for resolution. The resolution of the issue then depends on the expertise of the technicians or developers and their experience in handling similar issues which limits the size, speed, and scale of the problems that can be resolved. One solution to this problem is to make relevant information tailored to the issue under investigation to be easily available. Objectives: The focus of the thesis is to improve turn around time of customer queries using recommendations and evaluate by defining metrics in comparison to existing workflow. As Artificial Intelligence applications can have a broad spectrum, we confine the scope with a relevance in software service and Issue Tracking Systems. Software support is a complicated process as it involves various stakeholders with conflicting interests. During the course of this literary work, we are primarily interested in evaluating different AI solutions specifically in the customer support space customize and compare them. Methods: The following thesis work has been carried out by making controlled experiments using different datasets and Machine learning models. Results: We classified Axis data and Bugzilla (eclipse) using Decision Trees, K Nearest Neighbors, Neural Networks, Naive Bayes and evaluated them using precision, recall rate, and F-score. K Nearest Neighbors was having precision 0.11, recall rate 0.11, Decision Trees had precision 0.11, recall rate 0.11, Neural Networks had precision 0.13, recall rate 0.11 and Naive Bayes had precision 0.05, recall rate 0.11. The result shows too many false positives and true negatives for being able to recommend. Conclusions: In this Thesis work, we have gone through 33 research articles and synthesized them. Existing systems in place and the current state of the art is described. A debug suggestion tool was developed in python with SKlearn. Experiments with different Machine Learning models are run on the tool and highest 0.13 (precision), 0.10 (f-score), 0.11 (recall) are observed with MLP Classification Neural Network.

APA, Harvard, Vancouver, ISO, and other styles

24

Valério, Rafael Mendonça. "Projecto de data mining e modelação preditiva na TAP Portugal : relatório de estágio." Master's thesis, Instituto Superior de Economia e Gestão, 2014. http://hdl.handle.net/10400.5/8342.

Full text

Abstract:

Mestrado em Gestão de Sistemas de Informação
Este relatório de estágio dá a conhecer a importância que as ferramentas de análise preditiva como o data mining e a modelação preditiva têm para o sistema de Business Intelligence de uma organização. No caso concreto, tem-se como objeto de trabalho o sistema de gestão de receitas de uma companhia aérea portuguesa que nele identificou uma oportunidade de otimização. O projeto realizado teve como foco a implementação de uma componente de data mining que permitisse antecipar a receita relativa a bilhetes comprados mas não utilizados, geradora de valores significativos para a companhia. Esta abordagem tem a finalidade de suportar a tomada de decisão em várias vertentes do negócio, como custos operacionais, calendarização de voos, ações de marketing, comunicação, de entre outros. Identificados alguns problemas ao nível da estruturação dos dados analisados, tecem-se considerações acerca do tratamento destes.
This internship report was written in order to underline the importance of predictive analysis tools like data mining and predictive modelling to a Business Intelligence system in an organization. The object of this particular study is a Portuguese airline's revenue management system for which an optimization opportunity was identified: the implementation of a data mining component allowing to predict revenue regarding purchased, yet unused, plane tickets, which represents significant financial value to the airline. This approach aims to support decision making along several business areas, such as operational costs, flight scheduling, marketing practices and communication, among others. Since structural data problems were acknowledged during the project, some observations regarding data processing arise.

APA, Harvard, Vancouver, ISO, and other styles

25

Rouch, Jérémy. "Modélisations des systèmes d'assistance à la réverbération régénératifs." Phd thesis, Ecole Centrale de Lyon, 2013. http://tel.archives-ouvertes.fr/tel-00959768.

Full text

Abstract:

Les systèmes d'assistance a la réverbération sont des dispositifs électroacoustiques installés dans les salles de spectacle pour moduler leur acoustique en fonction du type de représentation qui s'y déroule. Afin de pouvoir dimensionner ces systèmes en amont de leur installation, ce travail s'intéresse au développement, à la mise en œuvre et à la mise à l'épreuve de modèles prévisionnels de l'effet de ses systèmes sur les caractéristiques acoustiques d'une salle, en se concentrant sur les systèmes dits régénératifs diagonaux. Le premier modèle présenté est basé sur une approche systémique exacte et sur l'utilisation de simulations numériques. Il s'agit d'un modèle dont le principe est déjà décrit dans la littérature spécialisée, mais auquel est intégré ici un algorithme de détermination automatique des paramètres de réglage d'un système d'assistance à la réverbération reproduisant la méthode manuelle. Parce que l'utilisation de simulations numériques impose une modélisation détaillée de la salle et un important temps de calcul, ce modèle n'est pas compatible avec la réactivité demandée lors d'une phase d'avant-projet. Dans cette optique, deux autres modèles bases sur les hypothèses de champ diffus et, par là même, plus rapides d'exécution, sont développés. L'un repose aussi sur une approche systémique exacte, mais utilise la théorie stochastique de l'acoustique géométrique des salles plutôt que des simulations numériques. L'autre repose sur une approche énergétique simple. Les confrontations de ces deux modèles avec celui reposant sur des simulations numériques sont exposées pour cinq salles, en considérant la prévision d'évolution de six indices acoustiques courants due à l'introduction d'un système d'assistance à la réverbération. Il en ressort que ces deux modèles aboutissent à des erreurs prévisionnelles comparables et que celles-ci sont équivalentes à celles des formules de Sabine ou d'Eyring ou de la théorie révisée des champs diffus de Barron et Lee appliquées dans une salle sans système. Parallèlement, l'étude et la prévision de l'effet d'un système d'assistance à la réverbération régénératif en augmentation du couplage de deux espaces mal couples au sein d'une même salle sont présentées. Il est montré que ce type d'utilisation permet effectivement une augmentation du couplage et que celle-ci peut être correctement abordée à partir d'un modèle énergétique développé ici. Il est aussi montré à partir de simulations numériques, que cette utilisation permet d'homogénéiser les caractéristiques acoustiques entre les deux espaces couplés.

APA, Harvard, Vancouver, ISO, and other styles

26

Huard, Malo. "Apprentissage et prévision séquentiels : bornes uniformes pour le regret linéaire et séries temporelles hiérarchiques." Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASM009.

Full text

Abstract:

Ce travail présente quelques contributions théoriques et pratiques à la prévision des suites arbitraires. Dans ce domaine, la prévision se déroule séquentiellement en même temps que l'apprentissage. À chaque étape, on ajuste le modèle sur les données passées afin de prévoir la prochaine observation. Le but de ce modèle est de faire les meilleures prévisions possibles, c'est-à-dire celles qui minimisent leurs écarts avec les observations. Les méthodes d'apprentissage séquentielles sont évaluées par leur regret, qui mesure à quelle point une stratégie est proche de la meilleure possible, qui est seulement connue une fois l'ensemble des données disponible. Un des résultats de cette thèse est d'étendre l'ensemble des stratégies auxquels on se compare lorsque l'on fait de la régression linéaire séquentielle. Nous avons adapté un algorithme existant en améliorant ses garanties théoriques pour lui permettre de se comparer à n'importe quelle combinaison linéaire constante sans restriction sur la norme de ses poids de mélange. Un deuxième travail a consisté à étendre les méthodes de prévisions séquentielles lorsque les données à prévoir sont hiérarchiquement organisées. Nous avons testé ces méthodes hiérarchiques sur deux applications pratiques, la prévision de consommation électrique des ménages et la prévision de ventes pour le e-commerce
This work presents some theoretical and practical contributions to the prediction of arbitrary sequences. In this domain, forecasting takes place sequentially at the same time as learning. At each step, the model is fitted on the past data in order to predict the next observation. The goal of this model is to make the best possible predictions, i.e. those that minimize their deviations from the observations, which are made a posteriori. Sequential learning methods are evaluated by their regret, which measures how close strategies are to the best possible, known only after all the data is available. In this thesis, we extend the set of weights vectors a method is compared to when doing sequential linear regression. We have adapted an existing algorithm by improving its theoretical guarantees allowing it to be compared to any constant linear combination without restriction on the norm of its mixing weights. A second work consisted in extending sequential forecasting methods when forcasted data is organized in a hierarchy. We tested these hierarchical methods on two practical applications, household power consumption prediction and demand forecasts in e-commerce

APA, Harvard, Vancouver, ISO, and other styles

27

Dimby, Solohaja Faniaha. "Détection d'outliers : modéllsation et prédiction : application aux données de véhicules d'occasion." Thesis, Paris 1, 2015. http://www.theses.fr/2015PA010025/document.

Full text

Abstract:

La société Autobiz édite et diffuse de l’information sur le secteur automobile. Cette thèse contribue à l’enrichissement de cette information et à une meilleure compréhension du marché de l’occasion par l’élaboration des modèles de prédiction du prix des véhicules et du délai de vente qui leur est associé. Nous avons eu à notre disposition une base de données réelles constituée d’annonces de sources diverses induisant un nombre considérable d’outliers. Ainsi, la première partie de travail s’est consacrée à la construction de méthodes de détection d’outliers incluant aussi bien de simples règles empiriques qu’un test statistique dont les propriétés asymptotiques ont été étudiées. Partant d’un état de l’art sur la prédiction des prix des véhicules d’occasion, il est apparu que les études existantes soulèvent le besoin de fonder une méthodologie d’analyse plus rigoureuse. Cette méthodologie a été développée dans un objectif de proposer des solutions automatisables et adaptées aux contraintes imposées par les experts. Nous faisons alors l’hypothèse que les prix des véhicules d’une même version se déprécient en fonction de l’âge et du kilométrage selon une forme qui lui est propre. La dernière partie du travail est dédiée à l’analyse des délais de vente. Dans un premier temps, nous caractérisons la variable associée aux délais de vente. Ensuite nous proposons une modélisation de cette variable par une régression à l’échelle d’un segment correspondant à l’arborescence marque-modèle-carrosserie-énergie en fonction des variables liées au kilométrage, au prix et à l’âge. Enfin, nous discutons de la possibilité de modéliser le nombre de véhicules vendus dans une période donnée selon une loi binomiale négative
Autobiz publishes information on the automotive sector. The subject of this the-sis is to give more tools for best understanding the used cars market by proposing modeling the price and the sale duration of vehicles. In our disposal we have a dataset consisted of used car advertisements automatically collected from the most popular website in France. Such data records often include outlying values. So, we need to start our analysis by considering outliers problem and we propose an outliers detector for univariate case for which we study asymptotic properties. Next, we develop a predicting model for used cars price. Although enumerable amount of works are stored in the literature we see that each of them lacks rigorous statistical foundations. We investigate the relationships between the price, the mileage, the age and others vehicle characteristics. More precisely we discuss how incorporate these variables in a model and compare different modeling approaches with the object to find the one best fitting the dataset and easy to implement. Expert’s opinions are minded at different stages of the model-building process. Next, we identify variables and how they affect the probability of a used vehicle’s sale from a list of explanatory variables related to price, mileage and age. In the sequel, we build a model allowing predicting the sale duration. Finally, we discuss about modeling sales of used cars by using the negative binomial distribution

APA, Harvard, Vancouver, ISO, and other styles

28

Hunter, Brandon. "Channel Probing for an Indoor Wireless Communications Channel." BYU ScholarsArchive, 2003. https://scholarsarchive.byu.edu/etd/64.

Full text

Abstract:

The statistics of the amplitude, time and angle of arrival of multipaths in an indoor environment are all necessary components of multipath models used to simulate the performance of spatial diversity in receive antenna configurations. The model presented by Saleh and Valenzuela, was added to by Spencer et. al., and included all three of these parameters for a 7 GHz channel. A system was built to measure these multipath parameters at 2.4 GHz for multiple locations in an indoor environment. Another system was built to measure the angle of transmission for a 6 GHz channel. The addition of this parameter allows spatial diversity at the transmitter along with the receiver to be simulated. The process of going from raw measurement data to discrete arrivals and then to clustered arrivals is analyzed. Many possible errors associated with discrete arrival processing are discussed along with possible solutions. Four clustering methods are compared and their relative strengths and weaknesses are pointed out. The effects that errors in the clustering process have on parameter estimation and model performance are also simulated.

APA, Harvard, Vancouver, ISO, and other styles

29

Chen, Jing-Hong, and 陳勁宏. "A SELECTIVE ENSEMBLE OF SALES PREDICTION MODELS." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/68014023536612232491.

Full text

Abstract:

碩士
中原大學
資訊管理研究所
96
Although the task of time series is full of noise and non-stationary, the value of its proper exploitation has attracted both researchers and practitioners. This thesis uses tools from the field of artificial intelligence (AI) such as the support vector machine (SVM) and the back propagation neural network (BPN) in order to predict the non-stationary movement of time series. More specifically, three ensemble strategies, i.e. the median based selective ensemble, the time-lag based ensemble, and the time-lag based selective ensemble are used. These three ensembles are designed to deal with the three problems, i.e. the low accuracy predicted by a single classifier due to the noise of data, not enough training samples as only data samples located near to the target sample are useful, the time lag problem of the traditional moving average (MA) approach. The first ensemble strategy handles the first problem successfully. The second and third ensemble strategies overcome the other problems. According to the experimental results from 50 small categories of products of the C company, the proposed ensemble strategies are able to deal with such three problems and therefore improve the prediction performance evaluated by the mean absolute percentage error (MAPE).

APA, Harvard, Vancouver, ISO, and other styles

30

Lai, Kuan-Hung, and 賴冠宏. "Automobile Sales Prediction Based On Case-Based Reasoning." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/t4pn5y.

Full text

Abstract:

碩士
國立勤益科技大學
資訊管理系
105
Recently, due to the signing of the WTO agreement and the ECFA agreement, and the close cooperation among car dealers, the car market in Taiwan has been turned from a closed market that is protected by the government into an open market. Hence, an effective way for car dealers to lower the cost is to make a prediction about car sales. Case-based analysis is a method for making predictions that does not require communication making among various professional fields, so it can raise efficiency of problem-solving. The method adopted in this study is regression analysis. Through the use of the regression analysis, the researcher tried to find out the environmental-economics factors that influenced the sales of the cars. Also, the method for the prediction of car sales that was based on case-based analysis was adapted. The data of these influential factors were optimized, and standardized. Then, they were combined with the case-based analysis to served as an adapted method of making predictions for car sales. The methods not only solved the problem of data with different measuring units, but also effectively solved the problem that the degree of similarity is influenced by the larger number when the numbers are extremely different and hence it could not show the influences of other factors. The result of this study indicated that the adapted method was superior to the traditional case-based analysis one and regression prediction analysis.

APA, Harvard, Vancouver, ISO, and other styles

31

Lu, Tz-ling, and 盧姿綾. "Application of Hilbert-Huang Transform in Sales Prediction Model." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/96804229869827453446.

Full text

Abstract:

碩士
東吳大學
企業管理學系
98
Sales prediction is an important issue for most enterprise in every industry. However, in the work of sales forecasting, the data of sales are usually affected by government’s policy, business cycle … etc. which make sales data include noises and instabilities. The noise of data will make time series model over fitting or under fitting and the instability will make it hard to construct a predicting model. For overcome the problem as above. This study uses the approaches of Hilbert-Huang transform (HHT), back-propagation neural network (BPN) and support vector regression (SVR). First, we use “Empirical Mode Decomposition” method of HHT to transform non-stationary and non-linear times series information into several “Intrinsic Mode Functions (IMFs)”. Second, we import IMFS by using BPN and SVR method to construct predicting model in order to reduce noise and instability. Finally, we use the sales data of six industries in Taiwan to test and verify the effectiveness of our method by compare with the data which are not transformed by HHT. Additionally, we also compare our method with Auto regressive Integrated Moving Average Models (ARIMA), Wavelet analysis and Independent Component Analysis. The results shows that our method is better than other model on predict errors.

APA, Harvard, Vancouver, ISO, and other styles

32

Chen, Mei-Feng, and 陳梅鳳. "A Comparison of Single and Multipal Sales Prediction Models." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/60699691796935021982.

Full text

Abstract:

碩士
中原大學
資訊管理研究所
96
Generally speaking, there are two groups of methods in the field of forecasting, i.e., traditional statistic analysis and artificial intelligence techniques. The first includes the regression analysis, correlation analysis, discriminate analysis, logit model, probit model, etc. The second includes decision tree, neural network, support vector machine, gray prediction, fuzzy theory, etc. Such methods may need different attributes for their specific application. Therefore, in literature, most forecasting tasks focus on specific attributes in their specific problem domains. In particular, they only use one or two forecasting methods to compare with one or two other methods after relevant attributes have been determined. The scale of such comparison is not wide enough. Thus, this thesis firstly uses multiple forecasting tools in artificial intelligence field, such as support vector machine (SVM) and neural network (NN). Secondly this thesis uses traditional statistical forecasting methods, such as simple linear regression (SLR), moving average (MA), ordinary least squares (OLS), and isotonic regression. Thirdly this thesis uses hybrid approaches by integrating artificial intelligence approaches with traditional statistical forecasting approaches. Finally, we also combine bagging and stacking ensemble learning techniques to get an improvement for our problem domain in the thesis. The standard of evaluation used in the thesis is the mean absolute percentage error (MAPE) that evaluates the most suitable forecaster from those (i.e., 16) forecasters in our specific problem domain which may provide more complete concept when similar forecasting task is performed in the future. The results in the thesis show OLS is the best model in the traditional statistical group, the NN forecaster is the best model in the artificial intelligence group, the SLR bagging forecaster is the best model in bagging group and the method which integrates NN bagging with stacking model is the best one in sixteen model.

APA, Harvard, Vancouver, ISO, and other styles

33

Xie, Jia-Ming, and 謝家銘. "Application of Bayesian Method for Chain Store Sales Prediction." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/av5u9a.

Full text

Abstract:

碩士
國立政治大學
統計學系
107
The prediction of sales is important. It is common to do regression analysis to predict sales for a store using its own data. However, for a chain with hundreds of stores, it may be possible to improve prediction accuracy and obtain more reasonable regression coefficients by combining data from different stores. We propose to achieve these goals by using two shrinkage methods: hierarchical Bayesian method and James-Stein estimator. We found that the shrinkage methods yield limited improvement when the regression coefficients in separate models are rather close. Moreover, the hierarchical method incorporated data from different stores and improve predictions, while James-Stein estimator did not improve much.

APA, Harvard, Vancouver, ISO, and other styles

34

Huang, Yih-Ling, and 黃奕綾. "The Commodities Sales Prediction: Cases of the CPC-LIFE Stores and Department Store Supermarket Sales Channels." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/bfz6he.

Full text

Abstract:

碩士
明道大學
企業高階管理碩士班
98
In Taiwan, the Convenience store and the Department store supermarket industry is being at the vigorous development stage, its marketing day by day innovates, the sales line of goods also presents the diversification, how to grasp the commodity the sales tendency, is in the sales management the great importance the work, is also grasps the realization of goods pulsation the important basis. The present paper conducts the sales predict research by the MSTR model in view of two kind of retail sales channels, the MSTR model content has the time series regression analytic method, the index smoothing procedures, the seller opinion synthesis method, and channel management rationalization parameter, and so on synthesis calculation models. This article carries on the market sales predict by the MSTR model; In the above two channels, uses the different forecast model in view of A~E kind and so on five kind of different category commodities, discusses its sales predict quantity, the accuracy. Because this research discovery these two kind of retail sales channel has the management multiplication, the product diversification, the customer status multi-densification, the purchase behavior to change and so on complexity, is not the sole forecast technique tallies sufficiently of sales predict all category commodity. The different sales predict method possibly is only suitable some kind of realization of goods predict that is unable to be suitable each kind of commodity the sales predict, even some kind of commodity its forecast deviation amount is very big. MSTR of model this research institute development, uses for to forecast that sales of this two kind of channel A~E five kind of commodity, the research discovered its predicted value and the actual value error are situated between compared to the MAPE value 0.10% to 8.43%, demonstrated that its can be effective, and widely forecasts this two channels, A~E and so on sales volumes of the five kind of commodity.

APA, Harvard, Vancouver, ISO, and other styles

35

Hung, Ming-Yang, and 洪明陽. "Sales Prediction Process Design and Model Building Under Promotion Period." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/12632190367021121438.

Full text

Abstract:

碩士
國立交通大學
經營管理研究所
100
In recent year, due to the strong competition and volatility of the market, firms must engage and cooperate with all the members in supply chain to sustain their competitive advantage by using supply chain management method. However, most of the methods have ignored the importance in optimization of information flow, which caused a serious impact to the whole supply chain, the “Long-Whip Effect”. The Long-whip effect indicates that the uncertainty and variation of demand information in supply chain will cause high inventory and high out-of-stock. This situation happens in promotion period often especially. For the demand will drive up in a very short period of time, long-whip effect happened if the supply chain cannot communicate to each other easily. Among all the supply chain management methods, collaborative planning, forecasting and replenishment (CPFR) is considered as the most effective way to deal with long-whip effect problem in supply chain. This study takes the engagement of a large supplier and a retailer in Taiwan as example, and aims to handle below three issues. First, this study re-designed the process in promotion period base on the basis of CPFR. Second, this study used multiple regression to discuss the key promotion factors that will affect sales quantity for 57 SKUs (stock keeping unit) that the supplier provided to the channel. Third, this study uses both multiple regression and Group Method Data Handling (GMDH) to predict sales quantity for each promotion period. With the total solution designed, the result has shown that the retailer has a better performance on out-of-stock rate, inventory reduction rate and order fulfillment rate that both side are eager to reach.

APA, Harvard, Vancouver, ISO, and other styles

36

SNENG, HSIEH PENG, and 謝朋昇. "Using Neural Network for the Sales Prediction of Domestic Cars." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/76234984798643561506.

Full text

Abstract:

碩士
大葉大學
工業工程與科技管理學系碩士在職專班
96
Automotive industry is one of the important manufacturing industries in Taiwan, and it is high related to other industries, and can push the related industries to upgrade. Thus, the development of automotive industry is the important index of country’s industry level, so it is also called the railway engine. The popular predict method of automotive industry is simple linear regression analysis. Although most of them take micro and market environment into consideration, the accuracy could be improved. Thus, to construct the prediction frame of the automotive market, including complete explanatory variables in order to enhance the predict method and be the reference for the enterprises and government proposing statistics, is the major motivation of this research. This study based on literary reviews and domestic automotive industry sales statistics during 1988~2006 BPNN proposes two conversion function, i.e. logsig and Tansig BPNN uses the previous as input to sixteen periods predict the current sales revenue The case is run through Matlab and MSE is used to judge the standard of internet termination After that,we summarize the actual predict performance. We compare the predict value to the estimation of linear regression analysis, and use MAPE as the evaluation of internet prediction, and then choose one with smallest errors as the best predict model. Conclusions show that the value of using BPNN with two conversion functions is better, and the value of Logsig is better than that of Tansig’s This proves that BPNN is the best method for this thesis, and is better than simple linear regression analysis.

APA, Harvard, Vancouver, ISO, and other styles

37

Chen, Jyun-Hong, and 陳俊宏. "Comparing different prediction techniques for sales forecasting in optical film industry." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/90026135518362505445.

Full text

Abstract:

碩士
健行科技大學
工業管理系碩士班
103
In this study, five different forecasting methods including naive forecast (NF), stepwise regression (SR), support vector regression (SVR), extreme learning machine (ELM) and back propagation neural network (BPN) are used to forecast sales amount of optical film industry. The monthly sales amount data collected from three optical film companies in Taiwan are used as experimental data to evaluate the performance of the five forecasting methods. The experimental results showed that SVR method can provide better forecasting results than that of the NF, SR, ELM and BPN. Thus, SVR is a promising technique for forecasting optical film sales amount.

APA, Harvard, Vancouver, ISO, and other styles

38

Lee, Yue-Hua, and 李玉華. "Application of Time Series and Neural Network Methods for Sales Prediction Model." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/r5w4a6.

Full text

Abstract:

碩士
國立中正大學
會計與資訊科技研究所
103
In hot summer, tea is one of the main sales in beverage, and drinking tea is becoming one of daily necessities today. Such a habit of drinking tea is quietly integrated into our lives. Because of changes in behavior of consumers in the market, convenience stores have become the main channel for selling tea. Today the speed of modern sales channel and the dividing profits by distributors have changed the manufacturers’ attitude to put effort on reducing the cost of sales in order to enhance their revenue. Therefore developing a production plan has become the important issue for the manufacturers. A well- developed production plan not only can reduce the production cost of tea, but also can enhance coordination between machines and employees, resulting in little waste of production resources. A good production plan relies on the accurate prediction of tea sales that is mainly dependent on consumer behaviors. However, the relationship among sale of products, consumers and other products is very complex. To address the issue about the prediction of tea sales, the traditional data analysis is not easy to predict the sales in the future market. Therefore manufacturers are looking forward to a better decision support systems for prediction. This study used tea sales data from point of sale (POS) in convenience store chain. We used data mining methodology to construct sales prediction model. The prediction method in the model is based on statistical time series moving average (MA) and autoregression integrated moving average (ARIMA). We also used neural network based on back-propagation network with changes in causal parameters for prediction models: One includes Granger causality test and the neural network ; another one includes Granger causality test, the neural network and more, the autoregressive (AR). We used actual sales data to assess these prediction models. Our results have shown that using Granger causality test plus autoregressive within back-propagation neural networks has the best prediction for these data.

APA, Harvard, Vancouver, ISO, and other styles

39

Ling-ChienHsu and 許凌倩. "IC Product Packaging and Sales Prediction Model Research--Using company Y as an example." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/19058904906630242557.

Full text

Abstract:

碩士
國立成功大學
工業與資訊管理學系專班
100
The semiconductor industry has been subject to tremendous flux amidst a period of severe global economic change. Therefore, how to predict economic change and plan for order shifting in an uncertain production environment are important topics for most semiconductor companies. As globalization is a defining trend of the future, prediction becomes one of the necessary managing tools for entrepreneurs. Most predictions nowadays are developed using nonlinear models, and employing artificial intelligence (AI) to engage in prediction is becoming more popular. Therefore, this research employs fuzzy-gray correlation analysis to sift factors which have higher correlations from different types of environmental factors, and find out the factor weight using the decision tree method of data mining, then placing those factors and weights into the back-propagation neural network (BPN) to engage in practices and predictions. This process is expected to increase the accuracy of predictions in the industry and improve order production predictions, and become a good reference for production prediction in the IC packaging industry.

APA, Harvard, Vancouver, ISO, and other styles

40

Wu, Cheng-Wei, and 吳晟瑋. "Constructing a Prediction Model for the Total Auction Sales of Flower Market in Taiwan." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/78086990151490566792.

Full text

Abstract:

碩士
國立交通大學
工業工程與管理系所
102
Auction is one of the commodity tradings. In Taiwan, there are many large auction markets which include fruit and vegetable market, hog market, fish market and flower market, etc. Although Taiwan only has five flower markets (namely, Taipei, Taichung, Tainan, Changhua and Kaohsiung), the total sales at auction between 1990 and 2000 was as high as NT $ 10 billions. Taiwanese flower market has a great contribution to Taiwan's economic development. As long as the total auction sales of various types of flowers can be predicted accurately, government can effectively advise regional flower farmers to plant the right amount and right kind of flowers to increase the profit. Therefore, the main objective of this study is to use auction information collected from five Taiwanese flower markets in recent years to construct prediction models of the auction amount and average auction price for certain type of flower, respectively. Then, integrating the predicted values of these two models to obtain the predicted value of total yearly auction sales of a certain flower. In this study, Autoregressive Integrated Moving Average (ARIMA) model is utilized to predict the auction amount of a certain flower such as carnations. As festivals (such as Mother's Day) have great impact on the total auction sales of certain types of flowers, the festival day is utilized as an explanatory variable to construct the prediction model of flower average auction price using Decomposition method in time series analysis. Finally, the total auction sales of a certain flower is obtained by multiplying its predicted auction amount and predicted average auction price. The results of the proposed method can be utilized for other kind of flowers.

APA, Harvard, Vancouver, ISO, and other styles

41

Omar, Hani, and 歐. 漢. 尼. "Data Mining for Sales Forecasting and Click-Through-Rate Prediction Based on Word Popularity." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/30930177440843525256.

Full text

Abstract:

博士
國立交通大學
資訊管理研究所
103
Internet technology has become a part of everyday life for retrieving data, contacting, entertainments, shopping, marketing, and some in the emerging business and developing world. Due to thousands of pages and services on the web, search engines are designed to search for information on the World Wide Web. The words of query are the main part in the retrieving results by search engines; and hence the word popularity is important to improve the correlated business for service providers. In this study, we first proposed a hybrid ARIMA and Back Propagation Neural Network for sales forecasting based on the popularity of article titles to enhance sales and operations planning. Publishing industries usually pick attractive titles and headlines for their stories to increase sales, since popular article titles and headlines can attract readers to buy or subscribe to magazines. The popularity of article titles are analyzed by using the search indexes obtained from Google search engine. We proposed a novel hybrid neural network model for sales forecasting based on the popularity of article titles, historical sales data, and the prediction result of Autoregressive Integrated Moving Average (ARIMA) forecasting method. Our proposed forecasting model is experimentally evaluated and the result shows that our proposed forecasting method outperforms conventional techniques which do not consider the popularity of title words. Second, we use the power of words of online advertisements, which impressed by search engines (where users add their queries for searching), to predict the users’ click-through rate (CTR) of advertisements. We use the important words in the queries which correlated to the advertisements and to boost the prediction performance. Also, we use the popularity of words to cope the cold-start problem when new users insert their query without having any knowledge about them using just their queries. Our proposed prediction model is evaluated and the result of the experiments shows that CTR prediction using word popularity outperform the prediction models without word popularity, and the same for cold start problem.

APA, Harvard, Vancouver, ISO, and other styles

42

Huang, Chih-Huang, and 黃智煌. "Ticket Sales Prediction of Entertainment Show Using Functional Data Clustering and Artificial Neural Network." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/w8w9u5.

Full text

Abstract:

碩士
國立臺灣科技大學
工業管理系
104
The sales performance of an entertainment show or concert tickets not only reflect profit of the business but also represents the popularity of the event. Predicting or forecasting the ticket sales performance before or during the ticket on sale is very important for the organization which hosts entertainment show event. In this research, a ticket sales prediction model was developed to predict the percentage of box office (ticket sales) of each price ranges based on the historical sales performance. In this research, a method called “Artificial Neural Network with Functional Data Clustering” (ANN_FDC) was proposed. Basically, the functional data clustering method is utilized to cluster show events by their ticket sale trajectory. Based on the clustering result, Artificial Neural Network (ANN) was developed for each ticket price range to predict the sales of the box office in terms of the percentage of ticket sold. This method is applied by using the first half of sale records to train the model for predicting the box office in the second half of sale periods. In this research, the 2010~2011 ticket sales data of Taipei area which usually exhibit pop music concerts was used as the testbed for evaluating the prediction model. The experimental results show the ANN_FDC can provide the better prediction and computational efficiency. This result can be further used for the study of the ticket marketing and sales promotion strategies.

APA, Harvard, Vancouver, ISO, and other styles

43

Wang, Fan-Ying, and 王汎嫈. "Sales Pipeline Prediction and Analysis by Using Logistic Regression: A Case of Fitness Equipment Business." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/mju462.

Full text

Abstract:

碩士
國立中興大學
高階經理人碩士在職專班
106
On big data analysis trend, how to use transaction data analysis to support problem solving in every corporate has drawn high attention of scholars and cooperate. Predictive analysis is one of often mentioned and discussed technology. It has been adopt in sales, operation, marketing, and risk analysis. Most of sales prediction analysis focus more on B2C, but studies of B2B transaction analysis are less discussed. Instead of discussion end users purchasing attribute on B2C predictive analysis, B2B transaction prediction care more on the win rate of each leads in pipeline. By more accurate transaction result prediction, cooperate would be able to better allocate limited resources, reduce stock inventory, improvement cash flow. This study is to use logistic regression to predict transaction result of sales pipeline final progress by using extracted data from sales pipeline management module of CRM system. The factors being booked in the sales pipeline system, such as customer, estimate revenue, payment term, warranty, lead time, sales channel have shown obvious effect to the independent factor. The predictive win rate by our calculation is much higher than actual situation, thus we could conclude that our study is in good direction and the predictive result could be a helpful tool to sales managers.

APA, Harvard, Vancouver, ISO, and other styles

44

Hwang, Jyi-Nan, and 黃吉南. "The Prediction Concerning the Sales Amount of Lottery Tickets and Expected Value of Money Award." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/91568395148725865266.

Full text

Abstract:

碩士
國立臺北大學
統計學系
91
During the initial issue of the lottery tickets, the public were overwhelmed by buying eagerness. The lottery craze was stunning. Two issues on the lottery tickets are addressed in this thesis. 1. The estimation of eagerness for buying the lottery tickets. 2. The test of randomness of the drawn winning numbers. For the first issue, the time series ARMA and ARCH models are established , in order to estimate the eagerness for buying the lottery tickets. The built models are then used to calculate the expectation of the winning amounts. For the second issue, under the null hypothesis that the six winning numbers randomly appear in each drawn , we propose several Goodness of fit tests. Simulation study is also given to illustrate our theoretical results. In this way the extracted statistics data are evaluated to see the level of fitness. We expect to learn whether the lottery is a fair game through the estimation of expectation value and the discussion of randomness, and thus come up with effective suggestions for the enthusiastic lottery players.

APA, Harvard, Vancouver, ISO, and other styles

45

Liu, Hsueh-Wei, and 劉學維. "Utilizing Sales Comparison Approach Combined with Data Mining Techniques to Construct Taipei Housing Price Prediction Models." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/92128692904391898181.

Full text

Abstract:

碩士
中原大學
資訊管理研究所
103
The people, banks, governments and courts have a problem to be solved for real estate appraisal that is "How much is a reasonable housing price?" However, an individual real estate appraisal for housing price will spend a lot of personnel, time and resources. Therefore, many related studies have been studied. In the past researches, most of them are sales comparison approach, hedonic price method, decision tree and neural network. For the sales comparison approach has subjective problems, but in selection method could reduce the data difference. Hedonic price method and neural network cannot join the subjective idea, but dealing with data difference may have a bias. Therefore, this research will combine the advantages of the above three methods. Use sale comparison approach selection method, and then use the hedonic price method or the data mining model. This study has the following advantages: (1) Reduce data difference and the possibility of influencing consequences. (2) Reduce the mean absolute percentage error (MAPE). (3) Be able to use computer assisted mass appraisal. (4) Objectivity. This study will not only investigate the above method but also study the new input values and different output values (Total price and unit price). The best result of this research method’s mean absolute percentage error (MAPE) is 23.23%. It beat the past 7.36%. Appraisal result is better than the past. The best selecting and modeling method select the past 16 seasons and within 0.5 km house data with the hedonic price method.

APA, Harvard, Vancouver, ISO, and other styles

46

Lee, Ta-Ching, and 李大經. "A Study of Performance Prediction and Design on Mobile Learning for Direct Selling Sales in Mainland China." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/14647660843406441625.

Full text

Abstract:

博士
淡江大學
管理科學學系博士班
104
The rise of smart phones from 2007 has stimulated various industries to develop mobile solutions to improve and solve past problems relating to the inaccessibility of information systems. This study used the direct marketing industry in Mainland China as the research subject. It aimed to understand the new opportunities and challenges encountered by the direct marketing industry in Mainland China and to propose mobile learning solutions to improve the relationship among extensive sales opportunities, skills of salespersons, and sales performance. Therefore, this study was divided into two phases: in the first phase, the study targeted the direct marketing industry in Mainland China through empirical research, using questionnaires after introducing mobile learning to analyze their validity and reliability with regarding to observing whether the mobile learning had a positive impact on sales performance. Research in this phase concluded that sales performance benefited from the introduction of mobile learning into the direct marketing industry. In the second phase, mobile learning solution was used as the means for implementing mobile solutions. Based on different user oriented scenario which was designed in the research, a prototype was designed and built for 300 salespersons. After those salespersons used and to be tested, the test results showed that sales performance for salespersons that used mobile learning solution was significantly higher compared to those who used traditional personal computers solution for sales efforts. This demonstrates that the use of mobile solutions in mobile learning improved sales quality in the direct marketing industry and increased the accuracy of merchandise sales.

APA, Harvard, Vancouver, ISO, and other styles

47

Ro, Mei-Her, and 羅美合. "The Prediction Model of the Sales and Price of Domestic Cars in Taiwan:the Application of State Space." Thesis, 1994. http://ndltd.ncl.edu.tw/handle/49408986285406894401.

Full text

APA, Harvard, Vancouver, ISO, and other styles

48

Chang, Chia-Lun, and 張嘉倫. "An Application of GM(1,1) on the Prediction of the Sales of Green Building Materials Market in Taiwan." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/98681149477211682038.

Full text

Abstract:

碩士
朝陽科技大學
企業管理系碩士班
96
Abstract In recent decades, continuous demanding consumption of natural resources and massive destruction of natural habitats have resulted in environmental degradation. Consequently, quest for sustainable living environment is no longer only voiced by environmental protection organizations but by general public as well. Benefited by the advancement of materials technology, the focus of construction materials has revolutionized from traditionally emphasized safe, comfort, and artistic to environmental and ecological friendly, sustainability, and health. Accompanied by the advocate of green building, more and more green building materials are developed and heavily promoted to the market. However, will demand soar? Or, the market expansion will be hindered by product quality, consumer acceptance, and other factors? Since the history of green building materials is relatively short, long-term accountable data are inaccessible. Thus, a model strong in analyzing small sample size would be required. As a result, grey prediction model GM (1,1) was selected for this study to discuss the future trends of green building materials. Result of this study on relevant selected products reveals an exciting message that there is indeed a growing market demand for green building materials. It shows that the demand of quarter three and four of 2008 will be 4,749 and 5,010 pieces respectively while the figure expects to grow to 5,283 and 5,573 pieces in quarter one and two of 2009 respectively. It is therefore viable for industry players to develop strategic plans to stay in competitive.

APA, Harvard, Vancouver, ISO, and other styles

49

WU, YI-HANG, and 吳易翰. "The Establishment and Prediction of Energy Sales Regression Model in Considering with Prosperity and Temperature for High Voltage Customer." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/mswppk.

Full text

Abstract:

碩士
國立高雄應用科技大學
電機工程系博碩士班
104
This thesis uses complex regression analysis method to establish customer’s load regression models, which consider economic indicators, temperature and rainfall. Furthermore, the proposed models are used to study the forecasting feasibility of the future energy sales and summer peak load demand. At first, this thesis uses least-squares techniques to derive regression models for considering economic indicators and temperature of 34 customer energy sales and total energy sales. Besides, the AMI high voltage customer demand data and system generating capacity for 24 hours are adopted to forecast summer peak load. Finally, the temperature sensitivity analysis is carried out to verify the result of service sector customers, energy intenstive customer and all high voltage customer energy sales. The above-mentioned data analysis tool is used by EViews software to achieve, in order to verify the feasibility of the research framework. The study found that through its energy intenstive customer and all high voltage customer aren’t sensitive to temperature, which shows forecasting model accuracy is low of only mixing with temperature and high voltage demand. So, mixing with high voltage demand data and system generating capacity for 24 hours to forecast peak load, which average error is ±0.87%. In the majority of its energy sales forecasting model of average error is ±3%. This result can be provided to power company as future reference.

APA, Harvard, Vancouver, ISO, and other styles

50

邱世梁. "A comparative study of prediction accuracy between Box-Jenkins and Pandit-Wu approaches:a case of retails sales of individual products." Thesis, 1986. http://ndltd.ncl.edu.tw/handle/04632521479892658173.

Full text

APA, Harvard, Vancouver, ISO, and other styles

Dissertations / Theses on the topic 'Prediction of sales'

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles