Inf in 'x'

Asked Nov 11 '20 at 15:14

Active Nov 11 '20 at 15:21

Viewed 208 times

Getting this error while trying to build linear regression model. But could see all variables are numeric. As I converted categorical into dummy variables few has 0's in it. When I don't apply log(sales) then it works fine. Facing issue only when i use log(sales). Can someone help me in this please.

[Code Snippet][1]

edited Nov 11 '20 at 15:21

asked Nov 11 '20 at 15:14

Priyanka Mahadevu

3

Do you have zero sales? `log(0)` returns `-Inf`. – Roland Nov 11 '20 at 15:20
Hi Priyanka, welcome to SO. please read https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example and avoid pasting images... – StupidWolf Nov 11 '20 at 15:22
I don't think it helps you if someone answers the question by posting an image for you :) so please help us to help you – StupidWolf Nov 11 '20 at 15:22
1

and you don't do all(is.na(...)) to find if there's any NAs. it should be any(is.na(..)) – StupidWolf Nov 11 '20 at 15:24

Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) : NA/NaN/Inf in 'x'

0 Answers0