变量选择是统计分析与建模领域中的一个重要议题,文章比较了Lasso方法和Adaptive lasso方法在高维混料模型中筛选变量的过程.通过实例验证, Adaptive lasso方法相较于Lasso方法具有较高的准确性且具相合性.通过Adaptive lasso方法得到一个满足Oracle性质的模型, 达到了降低成本、提高效益的目的.
Abstract
Variable selection is an important topic in the field of regression analysis. In this paper, we compare processes of variable selection for Lasso and Adaptive lasso with high-dimensional mixture model. Adaptive lasso can be more accurate and consistent lasso method. A model, which satisfies Oracle property and is attained by Adaptive lasso, can achieve the purpose of reducing costs and increasing benefit.
关键词
高维混料模型 /
变量选择 /
Lasso /
Adaptive lasso
{{custom_keyword}} /
Key words
high-dimensional mixture model /
variable selection /
Lasso /
Adaptive lasso
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] MALLOWS C L. Some comments on Cp[J]. Technometrics, 1973, 15(4):661-675.
[2] AKAIKE H. Information theory and an extension of the maximum likelihood principle[J]. International Symposium on Information Theory, 1973, 1:610-624.
[3] SCHWARZ G. Estimating the dimension of a model[J]. Annals of Statistics, 1978, 6:461-464.
[4] 曾津,周建军.高维数据变量选择方法综述[J].数理统计与管理,2017,36(4):678-692.
[5] TIBSHIRANI R. Regression shrinkage and selection via the lasso[J]. Journal of the Royal Statistical Society, 1996, 58:267-288.
[6] ZOU H. The adaptive lasso and its oracle properties[J]. Journal of the American Statistical Association,2006,101:1418-1429.
[7] 闫湛,张崇岐.混料试验设计的变量选择[J].数理统计与管理,2016,35(5):786-793.
[8] 张崇岐,闫湛.混料试验设计变量选择AIC准则研究[J].广州大学学报(自然科学版),2016,15(2):21-24.
[9] 李光辉,张崇岐.具有复杂约束混料试验的渐近D-最优设计[J].应用概率统计,2017,33(2):203-220.
[10]李光辉,张崇岐.混料试验的拟分量变换设计[J].应用数学学报,2017,40(5):734-751.
[11]EFRON B, HASTIE T, JOHNSTONE I. Least angle regression[J]. Annals of Statistics, 2004, 32(2):407-451.
[12]张建军, 付建武, 刘法锦. 混料设计优化复方双梅含片处方[J]. 中成药, 2011, 33(11):1908-1911.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金资助项目(11671104)
{{custom_fund}}
PDF(1454 KB)
