tag:blogger.com,1999:blog-1275149608391671670.post2366833583345384342..comments2023-09-28T06:13:40.704-04:00Comments on SAS and R: Example 8.14: generating standardized regression coefficientsKen Kleinmanhttp://www.blogger.com/profile/09525118721291529157noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-1275149608391671670.post-52693687321958087172017-07-04T11:21:17.884-04:002017-07-04T11:21:17.884-04:00I have a question that is and R question and a st...I have a question that is and R question and a statistical question:<br />I am analysing sales of a retailer. These sales are related to some vars: var1, var2, var3.., varN<br />Most of the vars are continuos. <br />I want to analyze the relationship between sales and the vars. I have made a linear regression with R:<br /><br />rg<-lm(sales ~ var1 + var2 + var3 + var4, data=sales_2017)<br />summary(rg)<br /><br />Now I want to know which is the most important variable in sales, and to know the percent of importance of each var. I am doing this (caret package):<br /><br />varImp(rg, scale = FALSE)<br />rsimp <- varImp(rg, scale = FALSE)<br />plot(rsimp)<br /><br />Is this a good method to obtain variables importance??, is good way in R?<br />Thanks in advance. Any advice will be greatly apreciated.<br /><br />JuanJ.V.https://www.blogger.com/profile/11172645377583674938noreply@blogger.comtag:blogger.com,1999:blog-1275149608391671670.post-1961727166945822342015-10-12T13:40:52.645-04:002015-10-12T13:40:52.645-04:00You can run "Make.Z()" in the QuantPsyc ...You can run "Make.Z()" in the QuantPsyc package to convert your data (then lm() would do this for you automatically).Nick Hortonhttps://www.blogger.com/profile/00242216324355342047noreply@blogger.comtag:blogger.com,1999:blog-1275149608391671670.post-15403568745216224762015-10-12T04:11:18.160-04:002015-10-12T04:11:18.160-04:00May I ask, how to get 95% confidence interval from...May I ask, how to get 95% confidence interval from standardized coefficients obtained from linear regression?Saznoreply@blogger.comtag:blogger.com,1999:blog-1275149608391671670.post-66863736247502500422013-06-20T17:12:47.791-04:002013-06-20T17:12:47.791-04:00Just curious, what is the rationale/support for th...Just curious, what is the rationale/support for the statement: "such an assessment ignores the confidence limits associated with each pairwise association"? Cheers!<br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-1275149608391671670.post-38153752631258248002010-11-17T12:30:16.370-05:002010-11-17T12:30:16.370-05:00I think it would make more sense to only standardi...I think it would make more sense to only standardize the continuous ones-- 2sd makes sense for them. I would leave the categorical covars as is, and also would not touch the outcome.Ken Kleinmanhttps://www.blogger.com/profile/09525118721291529157noreply@blogger.comtag:blogger.com,1999:blog-1275149608391671670.post-16433440458009560452010-11-16T14:28:23.100-05:002010-11-16T14:28:23.100-05:00Regarding the interpretation problem at the end, A...Regarding the interpretation problem at the end, Andrew Gelman makes a compelling argument for standardizing variables by 2 standard deviations so that the variance is similar to a binary variable (provided p is not too far from 0.5):<br /><br />http://onlinelibrary.wiley.com/doi/10.1002/sim.3107/abstract<br /><br />The arm package implements a standardize() function that appears to work similarly to lm.beta.Mark M. Fredricksonhttps://www.blogger.com/profile/13153103517132920757noreply@blogger.com