In this paper, we propose an approach to conduct partial sufficient dimension reduction with heavily unbalanced categorical predictors. For this, we consider integrated categorical predictors and investigate certain conditions that the integrated categorical predictor is fully informative to partial sufficient dimension reduction. For illustration, the proposed approach is implemented on optimal partial sliced inverse regression in simulation and data analysis.
Sufficient dimension reduction in multivariate regressions with categorical predictors, Computational Statistics & Data Analysis, 2013, 63, 139
Advances in seeded dimension reduction: Bootstrap criteria and extensions, Computational Statistics & Data Analysis, 2013, 60, 70
Generalized Partially Double-Index Model: Bootstrapping and Distinguishing Values, Communications for Statistical Applications and Methods, 2015, 22, 3, 305
Chiaromonte, F., Cook, R. D. and Li, B. (2002). Dimension reduction with categorical predictors, Annals of Statistics, 30, 475-497.
Cook, R. D. (1998). Regression Graphics, Wiley, New York.
Cook, R. D. and Li, B. (2002). Dimension Reduction for the Conditional Mean, Annals of Statistics, 30, 455-474.
Li, B., Cook, R. D. and Chiaromonte, F. (2003). Dimension reduction for the conditional mean in regressions with categorical predictors, Annals of Statistics, 31, 1636-1668.
Nierenberg, W. D., Stukel, A. T., Baron, A. J., Dain, J. B. and Greenberg, E. R. (1989). The skin cancer prevention study group, determinants of plasma levels of beta-carotene and retinol, American Journal of Epidemiology, 130, 511-521.
Wen, X. and Cook, R. D. (2007). Optimal sufficient dimension reduction in regressions with categorical predictors, Journal of Statistical Planning and Inference, 137, 1961-1978.
Yin, X. and Cook, R. D. (2002). Dimension reduction for the conditional kth moment in regression, Journal of Royal Statistical Society, Series B, 64, 159-175.
Yoo, J. K. (2009). Partial moment-based dimension reduction, Statistics and Probability Letters, 79, 450-456.