臺灣博碩士論文加值系統

Functional regression tree is used to predict the response variable which is a random function taking values in some infinite dimension. A method has been proposed that adapts the usual exhaustive search method for functional responses. We show that the method has selection bias toward variables with more split points. A selection is therefore designed to eliminate variable selection bias. It controls bias through utilizing chi-squared test for two-way contingency tables.

Functional regression tree is used to predict the response variable which is a random function taking values in some infinite dimension. A method has been proposed that adapts the usual exhaustive search method for functional responses. We show that the method has selection bias toward variables with more split points. A selection is therefore designed to eliminate variable selection bias. It controls bias through utilizing chi-squared test for two-way contingency tables.

1 Introduction 2
2 Variable selection 4
2.1 Exhaustive search method.................... 4
2.2 Proposed method............................. 5
3 Simulations 8
3.1 Selection bias.............................. 8
3.2 Power....................................... 10
4 Conclusion 14
Bibliography 15

Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C.J. (1984). Classification and Regression Trees. Wadsworth, Belmont.
Chaudhuri, D., Huang, M.-C., Loh, W.-Y., and Yao, R. (1994). Piesewise-polynomial regression trees. Statistica Sinica, 4:143-167.
Chaudhuri, D., Lo, W.-D., Loh, W.-Y., and Yang, C.-C. (1995). Generalized regression trees. Statistica Sinica, 5:641-666.
Csiszar, I. (1967). Information-type measures of difference of probability distributions and indirect observations. Studia Scientiarum Mathematicarum Hungarica, 2:299-318.
De''ath, G. (2002). Multivariate regression trees: a new technique for modeling species-environment relationships. Ecology, 83:1105-1117.
Grosjean, P., Picheral, M., Warembourg, C., and Gorsky, G. (2004). Enumeration, measurement, and identification of net zooplankton samples using the ZOOSCAN digital imaging system. ICES Journal of Marine Science, 61:518-525.
Hsiao, W.-C. and Shih, Y.-S. (2007). Splitting variable selection for multivariate regression trees. Statistics & Probability Letters, 77:265-271.
Larsen, D. R. and Speckman, P. L. (2004). Multivariate regression trees for analysis of abundance data. Biometrics, 60:543-549.
Lee, S. K. (2005). On generalized multivariate decision tree by using GEE. Computational Statistics & Data Analysis, 49:1105-1119.
Loh, W.-Y. (2002). Regression trees with unbiased variable selection and interaction detection. Statistica Sinica, 12:361-386.
Morgan, J. N. and Sonquist, J. A. (1963). Problems in analysis of survey data, and a proposal. Journal of the American Statistical Association, 58:415-434.
Nerini, D. and Ghattas, B. (2007). Classifying densities using functional regression trees: Applications in oceanology. Computational Statistics & Data Analysis, 51:4984-4993.