tag:blogger.com,1999:blog-7958828565254404797.post4976336709238930545..comments2019-09-22T07:28:56.135-07:00Comments on ListenData: Detecting and solving problem of OutlierDeepanshu Bhallahttp://www.blogger.com/profile/09802839558125192674noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-7958828565254404797.post-61080579384239975612018-02-26T09:10:08.707-08:002018-02-26T09:10:08.707-08:00There is nice explanation on Wikipedia under artic...There is nice explanation on Wikipedia under article "Interquartile range":<br />https://en.wikipedia.org/wiki/Interquartile_rangeAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-7958828565254404797.post-82751014126366317182016-09-11T20:54:01.135-07:002016-09-11T20:54:01.135-07:00Thanks for wonderful article!! I love all articles...Thanks for wonderful article!! I love all articles that you post.<br />Can you explain one we always look for Q1- 1.5 QTR/Q3+1.5QTR? I mean why 1.5 factor we are using not 1,2,2.5novice playernoreply@blogger.comtag:blogger.com,1999:blog-7958828565254404797.post-76398737189426619562016-07-10T10:58:55.714-07:002016-07-10T10:58:55.714-07:00WOE is more common for binary dependent variable. ...WOE is more common for binary dependent variable. It was originated from the logistic regression technique. You can use percentile capping (aka Winsorization) for treating outliers in linear regression. It can also be used in logistic regression. It is not restricted to linear regression.Deepanshu Bhallahttps://www.blogger.com/profile/09802839558125192674noreply@blogger.comtag:blogger.com,1999:blog-7958828565254404797.post-81667086229243990592016-07-10T05:50:58.967-07:002016-07-10T05:50:58.967-07:00WOE is used only in the logistic regression only ?...WOE is used only in the logistic regression only ? Can you please confirm which outlier treatment used for the linear regression only <br />Sandeep Phogathttps://www.blogger.com/profile/05718379449984928281noreply@blogger.comtag:blogger.com,1999:blog-7958828565254404797.post-75143853255173517852016-06-27T11:21:18.334-07:002016-06-27T11:21:18.334-07:00Outlier classes are grouped with other categories ...Outlier classes are grouped with other categories based on Weight of Evidence (WOE). For example, you have a income variable and outlier values are more than 500 million dollars. These values would be grouped to a class of (let's say 250-500 million dollars). Later, instead of using the raw values, we would be using WOE scores of each classes. Hope it helps. Thanks!Deepanshu Bhallahttps://www.blogger.com/profile/09802839558125192674noreply@blogger.comtag:blogger.com,1999:blog-7958828565254404797.post-49960357833658870462016-06-20T20:10:06.927-07:002016-06-20T20:10:06.927-07:00How WOE is used to detect and solve outlier proble...How WOE is used to detect and solve outlier problem?Abhinav Srivastavahttps://www.blogger.com/profile/05675515402356169746noreply@blogger.com