To see the other types of publications on this topic, follow the link: Multivariate Adaptive Regression Splines.
Dissertations / Theses on the topic 'Multivariate Adaptive Regression Splines'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 34 dissertations / theses for your research on the topic 'Multivariate Adaptive Regression Splines.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Kriner, Monika. "Survival Analysis with Multivariate adaptive Regression Splines." Diss., lmu, 2007. http://nbn-resolving.de/urn:nbn:de:bvb:19-73695.
Full text
2
Stevens, James G. "An investigation of multivariate adaptive regression splines for modeling and analysis of univariate and semi-multivariate time series systems." Thesis, Monterey, California. Naval Postgraduate School, 1991. http://hdl.handle.net/10945/26601.
Full text
3
Yazici, Ceyda. "A Computational Approach To Nonparametric Regression: Bootstrapping Cmars Method." Master's thesis, METU, 2011. http://etd.lib.metu.edu.tr/upload/12613708/index.pdf.
Full textAbstract:
Bootstrapping is a resampling technique which treats the original data set as a population and draws samples from it with replacement. This technique is widely used, especially, in mathematically intractable problems. In this study, it is used to obtain the empirical distributions of the parameters to determine whether they are statistically significant or not in a special case of nonparametric regression, Conic Multivariate Adaptive Regression Splines (CMARS). Here, the CMARS method, which uses conic quadratic optimization, is a modified version of a well-known nonparametric regression model, Multivariate Adaptive Regression Splines (MARS). Although performing better with respect to several criteria, the CMARS model is more complex than that of MARS. To overcome this problem, and to improve the CMARS performance further, three different bootstrapping regression methods, namely, Random-X, Fixed-X and Wild Bootstrap are applied on four data sets with different size and scale. Then, the performances of the models are compared using various criteria including accuracy, precision, complexity, stability, robustness and efficiency. Random-X yields more precise, accurate and less complex models particularly for medium size and medium scale data even though it is the least efficient method.
4
Stull, Kyra Elizabeth. "An osteometric evaluation of age and sex differences in the long bones of South African children from the Western Cape." Thesis, University of Pretoria, 2013. http://hdl.handle.net/2263/40263.
Full textAbstract:
The main goal of a forensic anthropological analysis of unidentified human remains is to
establish an accurate biological profile. The largest obstacle in the creation or validation of
techniques specific for subadults is the lack of large, modern samples. Techniques created for
subadults were mainly derived from antiquated North American or European samples and thus
inapplicable to a modern South African population as the techniques lack diversity and ignore
the secular trends in modern children. This research provides accurate and reliable methods to
estimate age and sex of South African subadults aged birth to 12 years from long bone lengths
and breadths, as no appropriate techniques exist.
Standard postcraniometric variables (n = 18) were collected from six long bones on 1380
(males = 804, females = 506) Lodox Statscan-generated radiographic images housed at the
Forensic Pathology Service, Salt River and the Red Cross War Memorial Children’s Hospital in
Cape Town, South Africa. Measurement definitions were derived from and/or follow studies in
fetal and subadult osteology and longitudinal growth studies. Radiographic images were
generated between 2007 and 2012, thus the majority of children (70%) were born after 2000 and
thus reflect the modern population.
Because basis splines and multivariate adaptive regression splines (MARS) are
nonparametric the 95% prediction intervals associated with each age at death model were
calculated with cross-validation. Numerous classification methods were employed namely linear,
quadratic, and flexible discriminant analysis, logistic regression, naïve Bayes, and random
forests to identify the method that consistently yielded the lowest error rates. Because some of
the multivariate subsets demonstrated small sample sizes, the classification accuracies were
bootstrapped to validate results. Both univariate and multivariate models were employed in the
age and sex estimation analyses.
Standard errors for the age estimation models were smaller in most of the multivariate
models with the exception of the univariate humerus, femur, and tibia diaphyseal lengths.
Univariate models provide narrower age estimates at the younger ages but the multivariate
models provide narrower age estimates at the older ages. Diaphyseal lengths did not demonstrate
any significant sex differences at any age, but diaphyseal breadths demonstrated significant sex
differences throughout the majority of the ages. Classification methods utilizing multivariate
subsets achieved the highest accuracies, which offer practical applicability in forensic
anthropology (81% to 90%). Whereas logistic regression yielded the highest classification
accuracies for univariate models, FDA yielded the highest classification accuracies for
multivariate models. This study is the first to successfully estimate subadult age and sex using an
extensive number of measurements, univariate and multivariate models, and robust statistical
analyses. The success of the current study is directly related to the large, modern sample size,
which ultimately captured a wider range of human variation than previously collected for
subadult diaphyseal dimensions.Thesis (PhD)--University of Pretoria, 2013.
gm2014
Anatomy
unrestricted
5
Pawar, Roshan. "Predicting bid prices in construction projects using non-parametric statistical models." [College Station, Tex. : Texas A&M University, 2007. http://hdl.handle.net/1969.1/ETD-TAMU-1464.
Full text
6
Lin, Yao. "An Efficient Robust Concept Exploration Method and Sequential Exploratory Experimental Design." Diss., Georgia Institute of Technology, 2004. http://hdl.handle.net/1853/4799.
Full textAbstract:
Experimentation and approximation are essential for efficiency and effectiveness in concurrent engineering analyses of large-scale complex systems. The approximation-based design strategy is not fully utilized in industrial applications in which designers have to deal with multi-disciplinary, multi-variable, multi-response, and multi-objective analysis using very complicated and expensive-to-run computer analysis codes or physical experiments. With current experimental design and metamodeling techniques, it is difficult for engineers to develop acceptable metamodels for irregular responses and achieve good design solutions in large design spaces at low prices. To circumvent this problem, engineers tend to either adopt low-fidelity simulations or models with which important response properties may be lost, or restrict the study to very small design spaces. Information from expensive physical or computer experiments is often used as a validation in late design stages instead of analysis tools that are used in early-stage design. This increases the possibility of expensive re-design processes and the time-to-market.
In this dissertation, two methods, the Sequential Exploratory Experimental Design (SEED) and the Efficient Robust Concept Exploration Method (E-RCEM) are developed to address these problems. The SEED and E-RCEM methods help develop acceptable metamodels for irregular responses with expensive experiments and achieve satisficing design solutions in large design spaces with limited computational or monetary resources. It is verified that more accurate metamodels are developed and better design solutions are achieved with SEED and E-RCEM than with traditional approximation-based design methods. SEED and E-RCEM facilitate the full utility of the simulation-and-approximation-based design strategy in engineering and scientific applications.
Several preliminary approaches for metamodel validation with additional validation points are proposed in this dissertation, after verifying that the most-widely-used method of leave-one-out cross-validation is theoretically inappropriate in testing the accuracy of metamodels. A comparison of the performance of kriging and MARS metamodels is done in this dissertation. Then a sequential metamodeling approach is proposed to utilize different types of metamodels along the design timeline.
Several single-variable or two-variable examples and two engineering example, the design of pressure vessels and the design of unit cells for linear cellular alloys, are used in this dissertation to facilitate our studies.
7
Yue, Yu. "Spatially adaptive priors for regression and spatial modeling." Diss., Columbia, Mo. : University of Missouri-Columbia, 2008. http://hdl.handle.net/10355/6059.
Full textAbstract:
Thesis (Ph. D.)--University of Missouri-Columbia, 2008.The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on August 3, 2009) Vita. Includes bibliographical references.
8
Schubert, Daniel Dice. "A multivariate adaptive trimmed likelihood algorithm /." Access via Murdoch University Digital Theses Project, 2005. http://wwwlib.murdoch.edu.au/adt/browse/view/adt-MU20061019.132720.
Full text
9
Kartal, Koc Elcin. "An Algorithm For The Forward Step Of Adaptive Regression Splines Via Mapping Approach." Phd thesis, METU, 2012. http://etd.lib.metu.edu.tr/upload/12615012/index.pdf.
Full textAbstract:
In high dimensional data modeling, Multivariate Adaptive Regression Splines (MARS) is a well-known nonparametric regression technique to approximate the nonlinear relationship between a response variable and the predictors with the help of splines. MARS uses piecewise linear basis functions which are separated from each other with breaking points (knots) for function estimation. The model estimating function is generated in two stepwise procedures: forward selection and backward elimination. In the first step, a general model including too many basis functions so the knot points are generatedand in the second one, the least contributing basis functions to the overall fit are eliminated. In the conventional adaptive spline procedure, knots are selected from a set of distinct data points that makes the forward selection procedure computationally expensive and leads to high local variance. To avoid these drawbacks, it is possible to select the knot points from a subset of data points, which leads to data reduction. In this study, a new method (called S-FMARS) is proposed to select the knot points by using a self organizing map-based approach which transforms the original data points to a lower dimensional space. Thus, less number of knot points is enabled to be evaluated for model building in the forward selection of MARS algorithm. The results obtained from simulated datasets and of six real-world datasets show that the proposed method is time efficient in model construction without degrading the model accuracy and prediction performance. In this study, the proposed approach is implemented to MARS and CMARS methods as an alternative to their forward step to improve them by decreasing their computing time
10
Sheppard, Therese. "Extending covariance structure analysis for multivariate and functional data." Thesis, University of Manchester, 2010. https://www.research.manchester.ac.uk/portal/en/theses/extending-covariance-structure-analysis-for-multivariate-and-functional-data(e2ad7f12-3783-48cf-b83c-0ca26ef77633).html.
Full textAbstract:
For multivariate data, when testing homogeneity of covariance matrices arising from two or more groups, Bartlett's (1937) modified likelihood ratio test statistic is appropriate to use under the null hypothesis of equal covariance matrices where the null distribution of the test statistic is based on the restrictive assumption of normality. Zhang and Boos (1992) provide a pooled bootstrap approach when the data cannot be assumed to be normally distributed. We give three alternative bootstrap techniques to testing homogeneity of covariance matrices when it is both inappropriate to pool the data into one single population as in the pooled bootstrap procedure and when the data are not normally distributed. We further show that our alternative bootstrap methodology can be extended to testing Flury's (1988) hierarchy of covariance structure models. Where deviations from normality exist, we show, by simulation, that the normal theory log-likelihood ratio test statistic is less viable compared with our bootstrap methodology. For functional data, Ramsay and Silverman (2005) and Lee et al (2002) together provide four computational techniques for functional principal component analysis (PCA) followed by covariance structure estimation. When the smoothing method for smoothing individual profiles is based on using least squares cubic B-splines or regression splines, we find that the ensuing covariance matrix estimate suffers from loss of dimensionality. We show that ridge regression can be used to resolve this problem, but only for the discretisation and numerical quadrature approaches to estimation, and that choice of a suitable ridge parameter is not arbitrary. We further show the unsuitability of regression splines when deciding on the optimal degree of smoothing to apply to individual profiles. To gain insight into smoothing parameter choice for functional data, we compare kernel and spline approaches to smoothing individual profiles in a nonparametric regression context. Our simulation results justify a kernel approach using a new criterion based on predicted squared error. We also show by simulation that, when taking account of correlation, a kernel approach using a generalized cross validatory type criterion performs well. These data-based methods for selecting the smoothing parameter are illustrated prior to a functional PCA on a real data set.
11
Ozmen, Ayse. "Robust Conic Quadratic Programming Applied To Quality Improvement -a Robustification Of Cmars." Master's thesis, METU, 2010. http://etd.lib.metu.edu.tr/upload/12612513/index.pdf.
Full textAbstract:
In this thesis, we study and use Conic Quadratic Programming (CQP) for purposes of operational research, especially, for quality improvement in manufacturing. In previous works, the importance and benefit of CQP in this area became already demonstrated. There, the complexity of the regression method Multivariate Adaptive Regression Spline (MARS), which especially means sensitivity with respect to noise in the data, became penalized in the form of so-called Tikhonov regularization, which became expressed and studied as a CQP problem. This was leading to the new method CMARSit is more model-based and employs continuous, actually, well-structured convex optimization which enables the use of Interior Point Methods and their codes such as MOSEK. In this study, we are generalizing the regression problem by including uncertainty in the model, especially, in the input data, too. CMARS, recently developed as an alternative method to MARS, is powerful in overcoming complex and heterogeneous data. However, for MARS and CMARS method, data are assumed to contain fixed variables. In fact, data include noise in both output and input variables. Consequently, optimization problem&rsquo
s solutions can show a remarkable sensitivity to perturbations in the parameters of the problem. In this study, we include the existence of uncertainty in the future scenarios into CMARS and robustify it with robust optimization which is dealt with data uncertainty. That kind of optimization was introduced by Aharon Ben-Tal and Arkadi Nemirovski, and used by Laurent El Ghaoui in the area of data mining. It incorporates various kinds of noise and perturbations into the programming problem. This robustification of CQP with robust optimization is compared with previous contributions that based on Tikhonov regularization, and with the traditional MARS method.
12
Tomek, Peter. "Approximation of Terrain Data Utilizing Splines." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2012. http://www.nusl.cz/ntk/nusl-236488.
Full textAbstract:
Pro optimalizaci letových trajektorií ve velmi malé nadmorské výšce, terenní vlastnosti musí být zahrnuty velice přesne. Proto rychlá a efektivní evaluace terenních dat je velice důležitá vzhledem nato, že čas potrebný pro optimalizaci musí být co nejkratší. Navyše, na optimalizaci letové trajektorie se využívájí metody založené na výpočtu gradientu. Proto musí být aproximační funkce terenních dat spojitá do určitého stupne derivace. Velice nádejná metoda na aproximaci terenních dat je aplikace víceroměrných simplex polynomů. Cílem této práce je implementovat funkci, která vyhodnotí dané terenní data na určitých bodech spolu s gradientem pomocí vícerozměrných splajnů. Program by měl vyčíslit více bodů najednou a měl by pracovat v $n$-dimensionálním prostoru.
13
Oubida, Regis Wendpouire. "Partitioning of multivariate phenotypes using regression trees reveals complex patterns of adaptation to climate across the range of black cottonwood (Populus trichocarpa)." Thesis, Virginia Tech, 2014. http://hdl.handle.net/10919/56619.
Full textAbstract:
Local adaptation to climate in temperate forest trees involves the integration of multiple physiological, morphological, and phenological traits. Latitudinal clines for the relevant component traits are frequently observed for species that have a north-south distribution, but these relationships do not account for climatic variation within a given latitudinal band, which may be reflected in adaptive traits. We used black cottonwood (Populus trichocarpa) as a model to characterize the interplay between geography, climate, and adaptation to abiotic factors. Twelve traits (height, diameter, volume index, crown diameter, number of branches, number of sylleptic branches, relative number of branches, Relative canopy depth, Bud set, Bud flush, cold index of injury, carbon isotope ratio) were measured in a range-wide sample of 124 P. trichocarpa genotypes grown in a common garden. Heritability's were moderate to high (0.24 to 0.55) and significant population differentiation (QST > 0.3) suggested adaptive divergence. When climate variables were taken as predictors and the 12 traits as response variables in a multivariate regression tree analysis, aridity (Eref) explained the most variation, with subsequent splits grouping individuals according to mean temperature of the warmest month, frost-free period (FFP), and mean annual precipitation (MAP). This grouping matches relatively well the splits using geographic variables as predictors: the northernmost groups (short FFP and low Eref) had the lowest growth performance, and the highest cold hardiness. The groups spanning the south of British Columbia (low Eref and intermediate temperatures) displayed an average growth and cold hardiness. The group from the coast of California and Oregon (high Eref and FFP) had the best growth performance and the lowest cold hardiness. The southernmost and high-elevated group (with High Eref and low FFP) performed poorly, had a low cold hardiness and a significantly lower WUE.Master of Science
14
Kidzinski, Lukasz. "Inference for stationary functional time series: dimension reduction and regression." Doctoral thesis, Universite Libre de Bruxelles, 2014. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/209226.
Full textAbstract:
Les progrès continus dans les techniques du stockage et de la collection des données permettent d'observer et d'enregistrer des processus d’une façon presque continue. Des exemples incluent des données climatiques, des valeurs de transactions financières, des modèles des niveaux de pollution, etc. Pour analyser ces processus, nous avons besoin des outils statistiques appropriés. Une technique très connue est l'analyse de données fonctionnelles (ADF).L'objectif principal de ce projet de doctorat est d'analyser la dépendance temporelle de l’ADF. Cette dépendance se produit, par exemple, si les données sont constituées à partir d'un processus en temps continu qui a été découpé en segments, les jours par exemple. Nous sommes alors dans le cadre des séries temporelles fonctionnelles.
La première partie de la thèse concerne la régression linéaire fonctionnelle, une extension de la régression multivariée. Nous avons découvert une méthode, basé sur les données, pour choisir la dimension de l’estimateur. Contrairement aux résultats existants, cette méthode n’exige pas d'assomptions invérifiables.
Dans la deuxième partie, on analyse les modèles linéaires fonctionnels dynamiques (MLFD), afin d'étendre les modèles linéaires, déjà reconnu, dans un cadre de la dépendance temporelle. Nous obtenons des estimateurs et des tests statistiques par des méthodes d’analyse harmonique. Nous nous inspirons par des idées de Brillinger qui a étudié ces models dans un contexte d’espaces vectoriels.
Doctorat en Sciences
info:eu-repo/semantics/nonPublished
15
Chang, Yi-Chen, and 張益城. "Multivariate Adaptive Regression Splines on Loss Reserve." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/73009419260625267884.
Full textAbstract:
碩士東吳大學
財務工程與精算數學系
104
Loss reserve is the insurer’s estimated total financial obligation for the claims that have not yet been paid, claims from policies written in the past. Estimate the loss reserve is one of the most important tasks for actuaries. Most popular methods in non-life loss reserving are Chain Ladder (CL) method, Expected Claims (EC) method, Bornhuetter-Ferguson (BF) method and Cape Cod (CC) method. These method not only easy to understand, has the beauty of simplicity, but also includes actuaries own experience. These methods are good for stable and short tailed type insurance, unknown for unstable and long tailed type contract. In this study we propose to use Multivariate Adaptive Regression Splines (MARS) to estimate the Loss Reserve. A set of real data is used. It is the claim data of general liability from one of Taiwan’s general insurance companies. General liability insurance covers variety of claims and, as result, severity is highly uncertain. We use this data to demonstrate the method of MARS estimation and verify tit’s accuracy. This data contains 20 accident quarters’ quarterly paid data for 20 development quarters, along with the earned premium. The 20 by 20 data are divided into two parts; the upper triangle is used for modeling purpose and the lower triangle for out-sample checking. We assume that the ultimate loss can be obtained by the fifth developing year, and calculate the severity at the end of fifth development year via MARS and four methods mentioned previously. Comparisons of the loss reserving and the actual severity are given: (1) MARS and CC methods are overestimate, slightly overestimate for the former and highly overestimate for the later method. (2) Other three methods are under estimate; BF method is slightly under estimate, results from EC and CL method are close, and CL method is the least favorable method. (3) In view of loss reserve should sufficient, but should not over conservative, for general liability insurance, we recommend MARS method over the other methods.
16
Lo, Hong Jie, and 羅宏杰. "Financial time series forecasting using multivariate adaptive regression splines." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/35714713071739540617.
Full text
17
Kriner, Monika [Verfasser]. "Survival analysis with multivariate adaptive regression splines / vorgelegt von Monika Kriner." 2007. http://d-nb.info/985556692/34.
Full text
18
Cheng, Chi-Heng, and 鄭啟亨. "Real Estate Price Estimates Using Evolutionary Fuzzy Multivariate Adaptive Regression Splines." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/60903943293709995625.
Full textAbstract:
碩士國立臺灣科技大學
營建工程系
103
Nowadays, the real estate prices are often estimates by the Real Estate Appraisers use of professional knowledge, experience, relevant information and precise judgment. But the different estimators will have different results. Nevertheless, this estimation is vulnerable due to human bias. To eliminate possible human bias made by appraisers. The research objective is to establish an Evolutionary Fuzzy Multivariate Adaptive Regression Splines (EFMARS) model, based on the hedonic pricing concept, to predict urban real estate price. Literature review summarizes 10 features that commonly show up for the hedonic pricing approach. The data collection targets at historical housing transactions in Taipei city Daan and Zhong Cheng district from Jan, 2014 to Dec, 2014. The total of 406 were collected. The study assessed model performance using the k-fold cross validation method and a stratified 10-fold cross validation approach. The results demonstrate that the proposed model reaches under 10% in MAPE. Compared with other modules, the result is also better than the Support Vector Machine (SVM), Neural Network (NN) and Regression, display model has a certain credibility.
19
Huang, Cheng-Feng, and 黃正鳳. "A Hybrid Classification Technique Using Neural Networks and Multivariate Adaptive Regression Splines." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/86700576097678634644.
Full textAbstract:
碩士輔仁大學
應用統計學研究所
90
Data mining is a very popular technique and has been widely applied in different areas these days. The artificial neural network is becoming a very popular alternative in prediction and classification task due to its associated memory characteristic and generalization capability. The objective of the proposed study is to explore the performance of data classification by integrating the artificial neural networks with the multivariate adaptive regression splines (MARS) approach. The rational under the analyses is firstly to use MARS in modeling the classification problem, then the obtained significant variables are then used as the input variables of the designed neural network model. To demonstrate the inclusion of the classification result from the multivariate adaptive regression splines would improve the classification accuracy of the networks, classification tasks are performed on one breast cancer data sets. As the results reveal, the proposed integrated approach outperforms the results using discriminant analysis, multivariate adaptive regression splines, classification and regression tree and artificial neural networks and hence provides an alternative in handling classification problems.
20
Yeh, Su-Mei, and 葉素美. "Using Neural Networks and Multivariate Adaptive Regression Splines in Exchange Rate Forecasting." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/25506987259006998879.
Full textAbstract:
碩士輔仁大學
應用統計學研究所
90
The development of Taiwan’s economy, as being an island in the Asian Pacific, heavily depends on international trade. And hence the profit margin of import/export traders is seriously impacted by foreign currency exchange rates. Fixed or stable exchange rate will definitely reduce the business risk of international traders. As Taiwan is relaxing the restriction of capital movement and trying to become the financial center of Asia Pacific region, a precise forecasting of the exchange rates is therefore an absolute necessity in implementing proper investment/financial decisions. The purpose of this research is to investigate the performance of three commonly discussed forecasting methods, backpropagation neural networks (BPN), multivariate adaptive regression splines (MARS), and ARIMA models, in exchange rate forecasting. The exchange rates analyzed in this study are five currencies against the US dollar, namely the New Taiwan (NT) dollar, Japanese Yen (Yen), British Pound (BP), Deutsche Mark (DM) and French Franc (FF). Analytic results demonstrate that the out-of-sample forecasts generated by the BPN models has better forecasting results than the other three models in most cases. However, the long training process in deciding the topology of BPN models is a factor needs to be carefully considered. On the other hand, the MARS model saves much more time than neural networks. It can provide reasonably good forecasting, except for the larger error of the weekly exchange rate on US dollar to Japanese Yen. Therefore, MARS provides an alternative in constructing exchange rate forecasting models with good forecasting precision and less model-building time.
21
I-Fei, Chen, and 陳怡妃. "Mining the Customer Retention Using Artificial Neural Networks and Multivariate Adaptive Regression Splines." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/05760834666953466719.
Full textAbstract:
碩士輔仁大學
管理學研究所
91
It is well know that, from the marketing’s point of view, the relevant cost of retaining an old customer is much lower than that of recruiting a new customer. On the other hand the customer profitability analysis has also gained serious attention since it is unwise to keep an unprofitable customer. As the competition between competitors is coming to a total conflicting stage and the pressure of lowering the marketing cost, more and more companies are seeking better ways to keep the profitable customers. Customer retention and customer profitability analyses both have been intensively studied due to the above-mentioned objectives. However, only very few literatures have tried to target these two subjects at the same time. In order to solve this drawback, this study tries to handle these two problems simultaneously aiming to identify the customers who are profitable but intend to leave. The built model can then provide useful information in implementing strategies in keeping profitable customers. The artificial neural network is becoming a very popular alternative in modeling customer retention and customer profitability problems due to its associated memory characteristic and generalization capability. However, the decision of network’s topology and the long training process has often long been criticized. The objective of the proposed study is to explore the performance of credit scoring by integrating the backpropagation neural networks (BPN) with multivariate adaptive regression splines (MARS) approach. To demonstrate the inclusion of the significant variables obtained from MARS would simplify the network structure and improve the classification accuracy of the designed neural network model, customer retention incorporating customer profitability tasks are performed on one health/fitness data set. As the results reveal, the proposed hybrid approach converges much faster than the conventional neural networks model. Moreover, the classification accuracies increase in terms of the proposed methodology and outperform traditional discriminant analysis, BPN and MARS approaches.
22
Ching-Yi, Chen, and 陳靜怡. "Forecasting the Unemployment Rate Using Artificial Neural Networks and Multivariate Adaptive Regression Splines." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/34225689236792658781.
Full textAbstract:
碩士輔仁大學
管理學研究所
91
After the presidential election in year 2000, Taiwan’s economy has suffered serious recession. The economic growth rate sharply decreased from 5.29% in the third quarter of year 2000 to 3.02% in the same period in year 2002. At the same time, the unemployment rate has also increased in an unbelievably fast pace. The total number of unemployed people has exceeded 500,000. The government is in dire need of proper strategies in reducing the unemployment rates. The artificial neural networks has become a useful tool for the forecasting of different economic behaviors. It is, however, also been criticized for its slow training process and unable to identify the relative importance of the input variables. The purpose of this research is to propose a two-stage unemployment forecasting model in integrating backpropagation neural network (BPN) and multivariate adaptive regression splines (MARS). Unlike past studies only use past observations or socioeconomic variables in building the forecasting model. This paper tries to incorporate both past observations and socioeconomic variables in building the model. MARS is first used to build the unemployment forecasting model with the obtained significant variables as the input nodes of the BPN model. In order to verify the effectiveness and predictive capability of the proposed two-stage modeling procedure, the monthly unemployment rate from September 1997 to October 2002 was used as an illustrative example. Empirical results indicate that the integrated approach provides significantly better forecasting results than solely using BPN, MARS, and ARIMA models. And the model with both past observations and socioeconomic variables as input variables also provides better results than those models only considering one type of variables
23
Huang, Ming-Hui, and 黃明輝. "Mining The Performance Of Mutual Funds Using Neural Networks And Multivariate Adaptive Regression Splines." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/84835063760647101602.
Full textAbstract:
碩士輔仁大學
金融研究所
90
Data mining is useful tools to discovery valid, novel, useful, new patterns or unknown relations in the data set or database. We also use data mining technology for finding out rules within database over the induction. Due to its applications to information systems, decision making, fraud detection, business failure prediction, database marketing, and lots of other applications, it has drawn serious attention from both academic researchers and practitioners. The purpose of this research is to investigate the performance of mutual fund bond using two commonly used data mining tools, artificial neural networks (ANNs) and multivariate adaptive regression splines (MARS). Several variables that may affect the performance mentioned in the literature, like the age of the bond fund, the portfolio of the bond fund, the scale of the bond fund, will be used to “predict” whether a particular bond will have the timing ability or not. In order to evaluate the classification capability of ANNs and MARS in mining the timing capability of mutual funds, historical date of 34 Taiwan bond funds from July 1999 to June 2001 will be used in this study. Analytic results demonstrate that MARS has better out-of-sample forecasts than ANNS in terms of average correct classification rates.
24
Yen, Yu-Ching, and 顏毓靜. "Investigating the Performance of Mutual Fund Using Neural Networks and Multivariate Adaptive Regression Splines." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/61451058087287063443.
Full textAbstract:
碩士輔仁大學
應用統計學研究所
90
Mutual fund has gained more and more attention since it is an efficient and alternative channel for public investors. As the amount invested in mutual fund has gone up steadily during the past decade and its high investment return, it has drawn serious attention from both academic researchers and practitioners. The purpose of this research is to build a classification model suitable for Taiwan’s mutual fund market and provide investors information about fund’s risk and performance. This research tries to investigate the performance of mutual fund in integrating artificial neural networks (ANNs) and multivariate adaptive regression splines (MARS). Several variables that may affect the performance mentioned in the literature, like the age of the bond fund, the portfolio of the bond fund, the scale of the bond fund, will be used to “predict” mutual funds’ risks and returns. MARS is first used to extract important variables which may influence mutual funds’ risks and returns. The obtained significant variables are then used as the input nodes of the designed neural network model to predict its risks and returns. In order to evaluate the classification capability of the proposed two-stage classification technique, historical date of 153 open stock funds are used in this study. Analytic results demonstrate that the integrated approach not only provides better classification accuracies but also saves lots of data processing time. It implies that the variables selected by MARS provide a better initial solution of the designed neural network model and hence provide better classification accuracies.
25
Lin, Chang-Jui, and 林長瑞. "Demand Forecasting of Chinese Tourists to Taiwan: Application of Multivariate Adaptive Regression Splines, the Artificial Neural Networks and Support Vector Regression." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/7njgg2.
Full textAbstract:
博士輔仁大學
商學研究所博士班
102
Since the starting of direct cross-strait transportation on August 2009, the number of Chinese visitors to Taiwan increased rapidly. By 2013, the number of Chinese visitors reached 2.87 million. Thus, Chinese visitors have become the most frequent visitors to Taiwan and the major source of foreign exchange earnings in tourism. This study applied three analysis model - multivariate adaptive regression splines (MARS), Artificial Neural Network (ANNs) and Support Vector Regression (SVR). Moreover, MARS combined with ANNs (MARS-ANNs) model and MARS combined with SVR (MARS-SVR) model were used to build the forecasting model for the number of Chinese visitors to Taiwan. MARS-SVR emerged as the best forecasting model and showed that the model combining with two analysis achieves better results compared to individual analysis tools. This study used MARS-SVR predictive model to forecast the number of Chinese visitors to Taiwan in 10 months - 7 months with forecasting error smaller than 5%, 3 months with forecasting error in between 5% to 10%. The accuracy of forecasting individual month is high; therefore, the result of MARS-SVR forecasting model can provide a reference for administrative authority and tourism industry to carry out construction program of tourism industry and plans for tourism activities. Six substantial forecasting variables were selected from eight forecasting variables by applying MARS analysis, which clearly indicated the high and low seasons. Therefore, administrative authority and tourism industry should pay attention to the difference in the number of Chinese visitors to Taiwan between high and low season. Accordingly, it is needed to increase flight times of cross-strait direct flights and additional tourist hotels to increase the transportation and accommodation capacity and serve more Chinese visitors. On the other hand, China’s gross domestic product (GDP), NTD to RMB exchange rate and the world oil price influence the number of Chinese visitors to Taiwan.
26
Liu, Shin-Yi, and 劉欣儀. "Artificial Neural Network and Multivariate Adaptive Regression Splines in Brand Valuation- Illustrative Examples of Global Top 100 Brands." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/43404779723269868329.
Full textAbstract:
碩士輔仁大學
管理學研究所
96
In recent years, the concept of intangible assets has become a very important research topic. As intangible assets often play roles in measuring the assets or brand value of companies, special attention should be paid to intangible assets in brand valuation. The purpose of this research is to forecast the brand value of global top 100 companies in integrating artificial neural networks and multivariate adaptive regression splines (MARS). The idea behind the model building approach is firstly using MARS to identify the significant input variables of the designed neural network model. Empirical results indicate that the proposed two-stage-hybrid MARS and neural networks forecasting method does have better forecasting capability in terms of the root mean squared error (RMSE) criterion. It therefore provides an alternative in exercising brand valuation. And in terms of the successful identification of the relationship within data, better business modeling can be found and implemented.
27
Li, Ai-Ti, and 李艾玓. "Enhancement of the Detection Capability of a SPC/EPC System Using Neural Networks and Multivariate Adaptive Regression Splines." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/14602642122419165868.
Full text
28
Yueh-Hsia, Lu, and 盧月霞. "Using back propagation network, multivariate adaptive regression splines, and autoregressive integrated moving average to design forecasting models for stock price." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/06346670509726255886.
Full textAbstract:
碩士輔仁大學
管理學研究所
96
Investing in stock market is the most popular and easiest way for investors. Everybody knows trading, but not all can make profit. As Taiwanese stock market doesn’t lie in the scope of efficient markets hypothesis, so investors can use reliable forecasting tools in predicting the trend and variation of stock prices. The purpose of this paper is to investigate the stock market forecasting capability of backpropagation neural network (BPN), multivariate adaptive regression splines (MARS), ARIMA, and hybrid MARS+BPN forecasting techniques. In order to evaluate the performance of the four proposed forecasting models, two public companies listed in Taiwan Stock Market are adopted as illustrative examples. The empirical results indicate that MARS and BPN provide better forecasting results in terms of several performance criteria. Besides, the obtained basis functions of MARS forecasting method can provide useful information for better investment decisions.
29
Tang, Pei-Wen, and 唐培文. "Integrating Neural Networks and Multivariate Adaptive Regression Splines in Price Prediction of Taiwan Stock Index Futures and Taiwan Top50 Tracker Fund." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/89695327956754179441.
Full textAbstract:
碩士輔仁大學
管理學研究所
93
This study predicts the opening prices and closing prices of TAIFEX index futures and Taiwan Top50 Tracker Fund ( Taiwan 50 ETF ) by the neural networks model and multivariate adaptive regression splines (MARS). The purpose of this study is to investigates if the two-stage hybrid modeling procedure with artificial neural networks and multivariate adaptive regression splines can find significant variables and then served as the input of the neural networks model. The empirical result shows the neural networks model using technical analysis indicators as inputs shows a better forecasting performance and the proposed hybrid approach outperforms the results only using artificial neural networks in both TAIFEX index futures and Taiwan Top50 Tracker Fund. In addition, the trading strategy of TAIEX index futures earns 118.06% annual return and the trading strategy of Taiwan Top50 Tracker Fund earns 11.54% annual return during the validation period.
30
Chen, Shing-Jin, and 陳興進. "Multivariate Regression Splines." Thesis, 1999. http://ndltd.ncl.edu.tw/handle/95792784455382660398.
Full textAbstract:
碩士國立交通大學
統計所
87
A multivariate spline of nonuniform smoothness condition is introduced and a multivariate B-spline basis of local support is developed for this spline space. Simulation results for these splines are provided.
31
Mondal, Anirban. "Bayesian Uncertainty Quantification for Large Scale Spatial Inverse Problems." Thesis, 2011. http://hdl.handle.net/1969.1/ETD-TAMU-2011-08-9905.
Full textAbstract:
We considered a Bayesian approach to nonlinear inverse problems in which the unknown quantity is a high dimension spatial field. The Bayesian approach contains a
natural mechanism for regularization in the form of prior information, can incorporate information from heterogeneous sources and provides a quantitative assessment of uncertainty in the inverse solution. The Bayesian setting casts the inverse solution as a posterior probability distribution over the model parameters. Karhunen-Lo'eve expansion and Discrete Cosine transform were used for dimension reduction of the
random spatial field. Furthermore, we used a hierarchical Bayes model to inject multiscale data in the modeling framework. In this Bayesian framework, we have shown that this inverse problem is well-posed by proving that the posterior measure is Lipschitz continuous with respect to the data in total variation norm. The need for multiple evaluations of the forward model on a high dimension spatial field (e.g. in the context of MCMC) together with the high dimensionality of the posterior, results in many computation challenges. We developed two-stage reversible jump MCMC method which has the ability to screen the bad proposals in the first inexpensive stage. Channelized spatial fields were represented by facies boundaries and
variogram-based spatial fields within each facies. Using level-set based approach, the shape of the channel boundaries was updated with dynamic data using a Bayesian
hierarchical model where the number of points representing the channel boundaries is assumed to be unknown. Statistical emulators on a large scale spatial field were introduced to avoid the expensive likelihood calculation, which contains the forward simulator, at each iteration of the MCMC step. To build the emulator, the original spatial field was represented by a low dimensional parameterization using Discrete Cosine Transform (DCT), then the Bayesian approach to multivariate adaptive regression spline (BMARS) was used to emulate the simulator. Various numerical results were presented by analyzing simulated as well as real data.
32
Chen, Chi-We, and 陳其瑋. "Multivariate Parallel Regression Splines." Thesis, 2000. http://ndltd.ncl.edu.tw/handle/09744226487078597430.
Full textAbstract:
碩士國立交通大學
統計所
88
Multivariate parallel spline of nonuniform smoothness condition is introduced and a multivariate parallel B-spline basis of local support is developed for this spline space.
33
Yeh, Shing-Hung, and 葉世弘. "Adaptive Group Lasso for Multivariate Linear Regression." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/90910161360611684952.
Full textAbstract:
碩士國立成功大學
統計學系碩博士班
97
In traditional statistical method, estimation and variable selection are almost discussed separately. LASSO (Tibshirani, 1996) is a new method for estimation in linear model, it can estimate parameters and variable selection simultaneously. But Lasso is inconsistent for variable selection, Adaptive Lasso (Zou 2006) overcomes these problems and enjoys the oracle properties. In linear regression when categorical predictors (factors) are present, the Lasso solution only selects individual dummy variables instead of whole factors. The group Lasso(Yuan and Lin 2006) overcomes these problems. Group lasso is a natural extension of lasso and selects variable in a grouped manner, group lasso suffers from estimation inefficiency and selection inconsistency. Adaptive Group Lasso (Wang and Leng 2006) show it’s estimator can be as efficient as oracle. We propose the adaptive group lasso for multivariate linear regression. In our study, the definition of grouped variable is different with the definition defined by formed study, which is regard one column of model matrix as a group. We consider one row of parametric matrix as one group for finding the significant variable on Y.
34
Hartman, Brian Matthew. "Bayesian Hierarchical, Semiparametric, and Nonparametric Methods for International New Product Di ffusion." Thesis, 2010. http://hdl.handle.net/1969.1/ETD-TAMU-2010-08-8294.
Full textAbstract:
Global marketing managers are keenly interested in being able to predict the sales
of their new products. Understanding how a product is adopted over time allows
the managers to optimally allocate their resources. With the world becoming ever
more global, there are strong and complex interactions between the countries in the
world. My work explores how to describe the relationship between those countries and
determines the best way to leverage that information to improve the sales predictions.
In Chapter II, I describe how diffusion speed has changed over time. The most
recent major study on this topic, by Christophe Van den Bulte, investigated new
product di ffusions in the United States. Van den Bulte notes that a similar study
is needed in the international context, especially in developing countries. Additionally,
his model contains the implicit assumption that the diffusion speed parameter
is constant throughout the life of a product. I model the time component as a nonparametric
function, allowing the speed parameter the
flexibility to change over time.
I find that early in the product's life, the speed parameter is higher than expected.
Additionally, as the Internet has grown in popularity, the speed parameter has increased.
In Chapter III, I examine whether the interactions can be described through
a reference hierarchy in addition to the cross-country word-of-mouth eff ects already
in the literature. I also expand the word-of-mouth e ffect by relating the magnitude
of the e ffect to the distance between the two countries. The current literature only applies that e ffect equally to the n closest countries (forming a neighbor set). This
also leads to an analysis of how to best measure the distance between two countries. I
compare four possible distance measures: distance between the population centroids,
trade
ow, tourism
ow, and cultural similarity. Including the reference hierarchy
improves the predictions by 30 percent over the current best model.
Finally, in Chapter IV, I look more closely at the Bass Diffusion Model. It is
prominently used in the marketing literature and is the base of my analysis in Chapter
III. All of the current formulations include the implicit assumption that all the
regression parameters are equal for each country. One dollar increase in GDP should
have more of an eff ect in a poor country than in a rich country. A Dirichlet process
prior enables me to cluster the countries by their regression coefficients. Incorporating
the distance measures can improve the predictions by 35 percent in some cases.