Optimal sufficient dimension reduction in regressions with categorical predictors

Though partial sliced inverse regression (partial SIR: Chiaromonte et al. [2002. Sufficient dimension reduction in regressions with categorical predictors. Ann. Statist. 30, 475–497]) extended the scope of sufficient dimension reduction to regressions with both continuous and categorical predictors, its requirement of homogeneous predictor covariances across the subpopulations restricts its application in practice. When this condition fails, partial SIR may provide misleading results. In this article, we propose a new estimation method via a minimum discrepancy approach without this restriction. Our method is optimal in terms of asymptotic efficiency and its test statistic for testing the dimension of the partial central subspace always has an asymptotic chi-squared distribution. It also gives us the ability to test predictor effects. An asymptotic chi-squared test of the conditional independence hypothesis that the response is independent of a selected subset of the continuous predictors given the remaining predictors is obtained.

[1]  Axel Ruhe,et al.  Algorithms for separable nonlinear least squares problems , 1980 .

[2]  Ker-Chau Li.  Sliced inverse regression for dimension reduction (with discussion) , 1991 .

[3]  S. Velilla.  Assessing the Number of Linear Components in a General Regression Problem , 1998 .

[4]  Thomas S. Ferguson,et al.  A Method of Generating Best Asymptotically Normal Estimates with Application to the Estimation of Bacterial Densities , 1958 .

[5]  R. Cook.  Save: a method for dimension reduction and graphics in regression , 2000 .

[6]  Ker-Chau Li,et al.  Sliced Inverse Regression for Dimension Reduction , 1991 .

[7]  Bruce G. Lindsay,et al.  Inference Functions and Quadratic Score Tests , 2003 .

[8]  M. Browne.  Asymptotically distribution-free methods for the analysis of covariance structures. , 1984, The British journal of mathematical and statistical psychology.

[9]  A. Shapiro.  Asymptotic Theory of Overparameterized Structural Models , 1986 .

[10]  D. Ruppert,et al.  Transformation and Weighting in Regression , 1988 .

[11]  S. Weisberg,et al.  Comments on "Sliced inverse regression for dimension reduction" by K. C. Li , 1991 .

[12]  Wolfgang Härdle,et al.  Sliced inverse regression for dimension reduction. Comments. Reply , 1991 .

[13]  R. Cook,et al.  Reweighting to Achieve Elliptically Contoured Covariates in Regression , 1994 .

[14]  R. Cook,et al.  Sufficient Dimension Reduction and Graphics in Regression , 2002 .

[15]  Ker-Chau Li,et al.  On almost Linearity of Low Dimensional Projections from High Dimensional Data , 1993 .

[16]  Bing Li,et al.  Dimension reduction for the conditional mean in regressions with categorical predictors , 2003 .