Timeline for Confidence interval for GLM or the maximum of a function?
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 15, 2022 at 23:01 | answer | added | kjetil b halvorsen♦ | timeline score: 3 | |
| May 15, 2022 at 23:00 | history | edited | kjetil b halvorsen♦ |
edited tags
|
|
| Jul 17, 2015 at 18:30 | history | edited | skan | CC BY-SA 3.0 |
added 5 characters in body
|
| Jul 17, 2015 at 18:22 | comment | added | skan | Maybe it's easier to run a small simulation with pertubation. I've added this idea as a second option. | |
| Jul 17, 2015 at 18:21 | history | edited | skan | CC BY-SA 3.0 |
added 241 characters in body
|
| Jul 17, 2015 at 18:04 | comment | added | skan | And my last question. When I have already calculated the deviation of x I wanted to use the x+-Z·S/sqrt(n) formula but this assumes normality. What should I use instead? | |
| Jul 17, 2015 at 17:59 | comment | added | skan | I'm thinking that the simplest method, calculating the error of x with the errors of b and c, doesn't take into account the errors of y. And I should do a weighted glm in order to weight the data with its experimental error. | |
| Jul 17, 2015 at 17:47 | comment | added | skan | Thanks, that was going to be related my next question, vcov solves it. The question aboout how to get the error associated to "y" still remains. If you create an answer joinning all your comments I could vote you. | |
| Jul 17, 2015 at 17:44 | comment | added | Dason | Your first question doesn't make any sense to me. And you would need to get the full covariance matrix for the predictors. You can't just use the standard errors of the estimates themselves because they are correlated. In R you would use the vcov function to get the covariance matrix from the model. | |
| Jul 17, 2015 at 17:40 | comment | added | skan | R glm shows the standard errors for the coefficient estimates. Should I use them or the standard deviations instead? And I also need to know the error for the Y, How can I get it? From the residuals deviance, using the "confint" or how? | |
| Jul 17, 2015 at 17:32 | comment | added | skan | So do you think the delta method (what I called the square root of the sum of derivatives squared...) would be appropiate to calculate the predictor's confidence interval ? And what if I prefer the likelihood methodology? Do you know of any link with examples o further information? | |
| Jul 17, 2015 at 17:12 | history | edited | skan | CC BY-SA 3.0 |
added 15 characters in body
|
| Jul 17, 2015 at 13:55 | history | tweeted | twitter.com/#!/StackStats/status/622042035684556800 | ||
| Jul 17, 2015 at 12:28 | comment | added | Dason | If you want to stay frequentist you could look up the delta method which will provide a way to get standard errors for transformations of your parameters | |
| Jul 17, 2015 at 11:50 | history | edited | skan | CC BY-SA 3.0 |
added 27 characters in body
|
| Jul 16, 2015 at 11:10 | history | asked | skan | CC BY-SA 3.0 |