Timeline for Additive property of the regression coefficients (slopes)
Current License: CC BY-SA 4.0
20 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
S Oct 16, 2023 at 18:30 | history | bounty ended | Kernel | ||
S Oct 16, 2023 at 18:30 | history | notice removed | Kernel | ||
Oct 16, 2023 at 18:27 | vote | accept | Kernel | ||
Oct 16, 2023 at 15:39 | history | edited | User1865345 | CC BY-SA 4.0 |
edited body
|
Oct 16, 2023 at 15:31 | answer | added | Sextus Empiricus | timeline score: 4 | |
Oct 16, 2023 at 13:53 | comment | added | Kernel | I just wanted to say that linear ordinary regression is the covariance divided by variance. I think we are now on the same page. | |
Oct 16, 2023 at 13:05 | answer | added | Steven Gubkin | timeline score: 4 | |
Oct 16, 2023 at 13:00 | comment | added | whuber♦ | One problem is the ambiguity of your language, because sums of squares of data are not the same as variances of underlying error distributions. | |
Oct 15, 2023 at 17:20 | comment | added | Kernel | I think that the ordinary linear regression "sum of the square formula” is also well expressed in covariance and variance matrix form. stat.cmu.edu/~cshalizi/mreg/15/lectures/13/lecture-13.pdf | |
Oct 15, 2023 at 16:51 | comment | added | whuber♦ | One particular formula for slopes involves sums of squares. That's not the same as "a function of variance and covariance." You might find it more fruitful to think in terms of the kind of regression you are doing. For instance, with ordinary least squares regression, the slopes are chosen to minimize the sum of squares of residuals (no variances involved!). | |
Oct 15, 2023 at 16:33 | comment | added | Kernel | Coefficients are the slopes. I updated my notions. X is the predictor. Regression is a function in variance and covariance. | |
Oct 15, 2023 at 16:31 | history | edited | Kernel | CC BY-SA 4.0 |
added 9 characters in body
|
Oct 15, 2023 at 15:57 | comment | added | whuber♦ | Your notation is unclear: could you please explain what you mean by "coefficients"? Are you trying to tell us that the ordinary regressions all have identical estimated slopes of $r,$ perhaps? Are you treating $x$ as the regressor or as the response in these regressions? Why would variance have anything to do with additivity of the responses? | |
Oct 15, 2023 at 15:36 | comment | added | Kernel | Let us make it simple, linear regression. | |
Oct 15, 2023 at 14:33 | comment | added | Steven Gubkin | What kind of regression? | |
S Oct 15, 2023 at 13:37 | history | bounty started | Kernel | ||
S Oct 15, 2023 at 13:37 | history | notice added | Kernel | Canonical answer required | |
Oct 14, 2023 at 17:57 | history | edited | Kernel |
edited tags
|
|
Oct 13, 2023 at 11:24 | history | edited | Shawn Hemelstrand | CC BY-SA 4.0 |
fixed formatting and added some tags and title fix
|
Oct 13, 2023 at 10:50 | history | asked | Kernel | CC BY-SA 4.0 |