Timeline for Interpretation of log transformed predictor and/or response
Current License: CC BY-SA 4.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 17, 2022 at 16:02 | history | edited | COOLSerdash | CC BY-SA 4.0 | Updated dead links. |
| Jun 3, 2019 at 17:38 | comment | added | AdamO | See my answer here about why this is wrong, and why exponentiating the coefficient and 95% CIs is the preferred approach. | |
| Apr 27, 2019 at 17:50 | comment | added | Nick Cox | Links are broken. | |
| Apr 27, 2019 at 17:50 | history | edited | Nick Cox | CC BY-SA 4.0 | edited body |
| Dec 29, 2017 at 18:30 | comment | added | AdamO | I may be confused. If you log-transform the outcome, you must re-exponentiate the coefficient to find the multiplicative difference. Interpretting it on the log scale only works as an approximation when the ratio is very close to 1. | |
| Oct 1, 2015 at 9:03 | comment | added | Bakaburg | So a DV ~ B1*log(IV) is a good model for zero bounded continuous dependent variable? | |
| Nov 3, 2014 at 16:44 | comment | added | Antouria | Example B: Outcome transformed log(DV) = Intercept + B1 * IV + Error "One unit increase in IV is associated with a (B1 * 100) percent increase in DV In this case, how do you do if you want 30 pourcent of DV reduction ? Thank you for your answer | |
| Sep 24, 2014 at 14:52 | comment | added | Ayalew A. | Do these interpretations hold regardless of the base of the logarithm? | |
| Nov 19, 2011 at 20:51 | history | edited | jthetzel | CC BY-SA 3.0 | typo |
| Nov 19, 2011 at 19:27 | history | edited | jthetzel | CC BY-SA 3.0 | typo |
| Nov 19, 2011 at 19:15 | history | answered | jthetzel | CC BY-SA 3.0 |