期刊名称：Practical Assessment, Research and Evaluation
印刷版ISSN：1531-7714
电子版ISSN：1531-7714
出版年度：2016
卷号：21
期号：7
出版社：ERIC: Clearinghouse On Assessment and Evaluation
摘要：Researchers and data analysts are sometimes faced with the problem of very small samples, where thenumber of variables approaches or exceeds the overall sample size; i.e. high dimensional data. In suchcases, standard statistical models such as regression or analysis of variance cannot be used, eitherbecause the resulting parameter estimates exhibit very high variance and can therefore not be trusted,or because the statistical algorithm cannot converge on parameter estimates at all. There exist analternative set of model estimation procedures, known collectively as regularization methods, whichcan be used in such circumstances, and which have been shown through simulation research to yieldaccurate parameter estimates. The purpose of this paper is to describe, for those unfamiliar with them,the most popular of these regularization methods, the lasso, and to demonstrate its use on an actualhigh dimensional dataset involving adults with autism, using the R software language. Results ofanalyses involving relating measures of executive functioning with a full scale intelligence test scoreare presented, and implications of using these models are discussed
