臺灣博碩士論文加值系統

Two selection methods were proposed for classification trees with
multivariate responses. Both methods declare that their variable
selection methods are unbiased. So, this makes us to study the
performance of the two methods. Simulation results show that the
method of Lee & Shih (2006) is better than the method of Hothorn,
Hornik & Zeileis (2006) in most parts of the simulations.

Two selection methods were proposed for classification trees with
multivariate responses. Both methods declare that their variable
selection methods are unbiased. So, this makes us to study the
performance of the two methods. Simulation results show that the
method of Lee & Shih (2006) is better than the method of Hothorn,
Hornik & Zeileis (2006) in most parts of the simulations.

1 Introduction 1
2 Methods of selecting variables 3
2.1 CT method . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 PT method . . . . . . . . . . . . . . . . . . . . . . . 4
3 Simulations 7
3.1 Split selection bias . . . . . . . . . . . . . . . . . . . 7
3.1.1 Independent components of Y versus independent
X0s . . . . . . . . . . . . . . . . . . . 8
3.1.2 Dependent components of Y versus independent
X0s . . . . . . . . . . . . . . . . . . . . . 9
3.1.3 Independent components of Y versus dependent
X0s . . . . . . . . . . . . . . . . . . . . . 9
3.1.4 Dependent components of Y versus dependent
X0s . . . . . . . . . . . . . . . . . . . . . 11
3.2 Selection power . . . . . . . . . . . . . . . . . . . . . 11
4 Conclusion 18

Breiman, L., Friedman, J. H., Olshen, R. A. & Stone, C. J. (1984). Classification
and Regression Trees, Wadsworth Publishing Co Inc.
Hothorn, T., Hornik, K. & Zeileis, A. (2006). Unbiased recursive partitioning: A
conditional inference framework, Journal of Computational and Graphical
Statistics 15: 651–674.
Kim, H. & Loh, W.-Y. (2001). Classification trees with unbiased multiway splits,
Journal of the American Statistical Association 96: 589–604.
Lee, T.-H. & Shih, Y.-S. (2006). Unbiased variable selection for classification trees
with multivariate responses, Computational Statistics and Data Analysis
51: 659–667.
Leisch, F. & Weingessel, A. (2007). Bindata: Generation of Artificial Binary
Data. R package version 0.9-14.
Loh, W.-Y. & Shih, Y.-S. (1997). Split selection methods for classification trees,
Statistica Sinica 7: 815–840.
Loh, W.-Y. & Vanichsetakul, N. (1988). Tree-structured classification via generalized
discriminant analysis, Journal of the American Statistical Association
83: 715–728.
Noh, H. G., Song, M. S. & Park, S. H. (2004). An unbiased method for constructing
multilabel classification trees, Computational Statistics and Data
Analysis 47: 149–164.
Quinlan, J. R. (1993). C4.5: Programs for Machine Learning, Morgan Kaufmann
Publishers Inc., San Mateo, California.
R Development Core Team (2007). R: A Language and Environment for Statistical
Computing, R Foundation for Statistical Computing, Vienna, Austria.
ISBN 3-900051-07-0.
*http://www.R-project.org
Siciliano, R. & Mola, F. (2000). Multivariate data analysis and modeling through
classification and regression trees, Computational Statistics and Data Analysis
32: 285–301.
Simonoff, J. S. (2003). Analyzing Categorical Data, Springer.
Strasser, H. &Weber, C. (1999). The asymptotic theory of permutation statistics,
Mathematical Methods of Statistics 8: 220–250.
Zhang, H. (1998). Classification trees for multiple binary responses, Journal of
the American Statistical Association 93: 180–193.