Variable Reduction in SAS by Using Weight of Evidence and Information Value

相似文件 换一批

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

Variable reduction is a crucial step for accelerating model building without losing potential predictive power of the
data. This paper provides a SAS macro that uses Weight of Evidence and Information Value to screen continuous,
ordinal and categorical variables based on their predictive power. The results can lend useful insights to variable
reduction. The reduced list of variables enables statisticians to quickly identify the most informative predictors for
building logistic regression models.


Variable Reduction in SAS by Using Weight of Evidence and Information Value.pdf (899.9 KB)

2016-2-19 12:09:56 上传

Variable Reduction in SAS by Using Weight of Evidence and Information Value

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：information Informatio formation Reduction informat continuous potential building identify provides

谢谢分享

http://support.sas.com/resources/papers/proceedings13/095-2013.pdf

顶下楼主和三楼