Timeline for 300k observations, significant results but very low adjusted r-squared
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 8, 2020 at 22:00 | history | edited | Firebug | CC BY-SA 4.0 |
edited tags
|
| Jul 12, 2019 at 0:43 | comment | added | Dave | Adjusted $R^2$ is just one way to guard against overfitting. Have you tried to train on 250k samples and find $R^2$ for the predictions of the 50k that your model did not see during training? | |
| May 24, 2019 at 17:57 | comment | added | Michael R. Chernick | I wasn't suggesting reducing the sample size. The larger the sample size the better. A low R$*2$ could indicate a poor model. The fact that it is statistically significantly different from 0 is useful information. | |
| May 24, 2019 at 4:00 | vote | accept | Deepan Das | ||
| May 24, 2019 at 3:43 | vote | accept | Deepan Das | ||
| May 24, 2019 at 4:00 | |||||
| May 24, 2019 at 3:41 | answer | added | benws | timeline score: 3 | |
| May 24, 2019 at 3:40 | vote | accept | Deepan Das | ||
| May 24, 2019 at 3:43 | |||||
| May 24, 2019 at 3:36 | answer | added | Frans Rodenburg | timeline score: 4 | |
| May 24, 2019 at 3:23 | comment | added | Deepan Das | I unfortunately cannot reduce the sample size; it has to be like this. Can the low adjusted r-squared owing to the large sample size cast any doubt on the significance of the variables? Much thanks for your response. | |
| May 24, 2019 at 3:17 | comment | added | Michael R. Chernick | This is not a strong model but because the sample size is so large you can detect a small positive R$^2$. | |
| May 24, 2019 at 3:10 | review | First posts | |||
| May 24, 2019 at 3:17 | |||||
| May 24, 2019 at 3:08 | history | asked | Deepan Das | CC BY-SA 4.0 |