0

I ran a specification with an outcome in $ terms, and a coefficient on a dummy variable was statistically significant. However, when I ran the exact same model with the outcome in log-dollar terms, the coefficient is not significant. Is this possible, or must there be a mistake?

user43940
  • 3
  • 2
  • 1
    This is a frequently asked question. You basically need to understand what a log model vs a non-log model is, so you should focus on the thread above. Following that, here are some more threads to read: [What if a log transformation wipes out significance in regression?](http://stats.stackexchange.com/q/79276/7290); & [Log transformed variable not significant, while variable itself is](http://stats.stackexchange.com/q/74855/7290). – gung - Reinstate Monica Apr 17 '14 at 22:12

1 Answers1

0

That is perfectly possible. In my experience, this is often because of outliers, who tend to be less important when a log-transform is applied. Having said that, please observe that if you use logs, you are basically estimating a different quantity, so it can very well be that significance is lost even though there are no outliers whatsoever.

coffeinjunky
  • 1,646
  • 1
  • 16
  • 22