A recursive partitioning tool for interval prediction Export

@article{2007_KrzanowskiHand, abstract = {Abstract The traditional approach to regression trees involves partitioning the space of predictor variables into subsets that optimise a function of the response variable(s), and then predicting future response values by a single-valued summary statistic in each subset. Our belief is that a prediction interval is of greater practical use than a predictive value, and that the criterion for the partitioning should be based on such intervals rather than on single values. We define four potential criteria in the case of a single response variable, discuss computational aspects of producing the partition, evaluate the criteria on both real and simulated data, and draw some tentative conclusions about their relative efficacies. The methodology is extended to the case of multiple response variables, and its viability is demonstrated by application to some further real data. The possibility of fitting distributions to within-subsets data is discussed, and some potential extensions are briefly outlined.}, author = {Krzanowski, Wojtek and Hand, David}, citeulike-article-id = {3083564}, citeulike-linkout-0 = {http://dx.doi.org/10.1007/s11634-007-0015-y}, day = {25}, doi = {10.1007/s11634-007-0015-y}, journal = {Advances in Data Analysis and Classification}, keywords = {cart, clustering, ic\_srf, qfrmc, regression, trees}, month = {December}, number = {3}, pages = {241--254}, posted-at = {2008-08-05 08:40:13}, priority = {0}, title = {A recursive partitioning tool for interval prediction}, url = {http://dx.doi.org/10.1007/s11634-007-0015-y}, volume = {1}, year = {2007} }

TY - JOUR ID - 2007_KrzanowskiHand L3 - citeulike-article-id:3083564 N2 - Abstract The traditional approach to regression trees involves partitioning the space of predictor variables into subsets that optimise a function of the response variable(s), and then predicting future response values by a single-valued summary statistic in each subset. Our belief is that a prediction interval is of greater practical use than a predictive value, and that the criterion for the partitioning should be based on such intervals rather than on single values. We define four potential criteria in the case of a single response variable, discuss computational aspects of producing the partition, evaluate the criteria on both real and simulated data, and draw some tentative conclusions about their relative efficacies. The methodology is extended to the case of multiple response variables, and its viability is demonstrated by application to some further real data. The possibility of fitting distributions to within-subsets data is discussed, and some potential extensions are briefly outlined. IS - 3 JF - Advances in Data Analysis and Classification EP - 254 TI - A recursive partitioning tool for interval prediction VL - 1 SP - 241 KW - cart KW - clustering KW - ic_srf KW - qfrmc KW - regression KW - trees AU - Krzanowski, Wojtek AU - Hand, David PY - 2007/12/25/ UR - http://dx.doi.org/10.1007/s11634-007-0015-y ER -