Timeline for Multiple linear regression with restricted ranges in R
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Jan 22, 2013 at 9:04 | history | edited | chl | CC BY-SA 3.0 |
corrected spelling; improved formatting
|
| S Jan 22, 2013 at 9:04 | history | suggested | rcs | CC BY-SA 3.0 |
corrected spelling; improved formatting
|
| Jan 22, 2013 at 7:47 | review | Suggested edits | |||
| Jan 22, 2013 at 9:04 | |||||
| Jan 21, 2013 at 13:02 | history | tweeted | twitter.com/#!/StackStats/status/293342917043560448 | ||
| Jan 21, 2013 at 12:36 | answer | added | Günal | timeline score: 1 | |
| Jan 21, 2013 at 12:16 | answer | added | IMA | timeline score: 1 | |
| Jan 21, 2013 at 11:48 | comment | added | Peter Flom | I fixed up the formatting, but do you really mean to have $C_0$ five times? This would be most unusual; I think you want $C_1$, $C_2$ etc. Also, you didn't include an intercept; was that deliberate? | |
| S Jan 21, 2013 at 11:45 | history | edited | Peter Flom | CC BY-SA 3.0 |
Improved formatting.
|
| S Jan 21, 2013 at 11:45 | history | suggested | Sven Hohenstein | CC BY-SA 3.0 |
Improved formatting.
|
| Jan 21, 2013 at 8:46 | review | Suggested edits | |||
| Jan 21, 2013 at 11:45 | |||||
| Jan 21, 2013 at 8:26 | comment | added | user19869 | no problem....linear model's a start...right now I get negative values for one of the coefficients, which I can't have. | |
| Jan 21, 2013 at 8:21 | comment | added | IMA | Oh I completely misunderstood you then. Sorry. May I ask: Is the linear model necessarily appropriate? What results do you get when estimating right now? | |
| Jan 21, 2013 at 8:17 | comment | added | user19869 | I'm trying to fit some biological data, and need to constrain the coefficients such that they are not nonsensical... | |
| Jan 21, 2013 at 8:07 | review | First posts | |||
| Jan 21, 2013 at 8:46 | |||||
| Jan 21, 2013 at 8:07 | comment | added | IMA | What is the reason to constrain the ranges? If it is just a technical question you should maybe ask on stackoverflow, since this is more of a stats theoretical site. If the reason for constraint is a inheret to your model, I'd advise you to consider a (semi)-logarithmic model depending on what you are doing. Cutting out data point entirely tends to be a really bad idea for inference. | |
| Jan 21, 2013 at 7:49 | history | asked | user19869 | CC BY-SA 3.0 |