LSIC: least squares fitting under covariate shift using the information criterion
LSIC consists of short programs written in S language. It calculates the least squares fitting using the optimal weight under the "covariate-shift." The newly developed information criterion is used for the optimization. Covariate-shift is the change of the distribution of the covariates in the regression; the distribution for the observed data can be different from that of the population for the evaluation of the fitting. The theory and the method are described in Shimodaira (2000).
Shimodaira, H. Improving predictive inference under covariate shift by weighting the log-likelihood function. Journal of Statistical Planning and Inference 90, 227-244 (2000).