Log in
Register|
yes_naresh ☆ 2009-11-26 11:45 (6054 d 11:11 ago) Posting: # 4394 Views: 5,432 |
|
|
Dear All, Description of problem, 1) There are some observations. 2) with 6 independent variables and 1 response (Y) 3) I am going to fit regression model. 4) And there, I want to use Monte carlo simulation, HOW? Kindly discuss any method which I can use other than regression model (If assumptions requirement does not fulfill). Regards Naresh |
|
Helmut ★★★   Vienna, Austria, 2009-11-26 15:32 (6054 d 07:25 ago) @ yes_naresh Posting: # 4395 Views: 4,451 |
|
|
Dear Naresh! ad 1) Hopefully a lot of observations (for 6 linear regressors >18). ad 2) Fine. That's called multiple regression. ad 4) From your last sentence I guess you think that (some) assumptions of parametric regression do no hold. Have you tried the more common approaches: transformations or weighting? If this doesn't help, you may opt for robust regression. You may try to minimize not the sum of squared residuals (like in OLS), but the sum of absolute residuals, or the median of residuals, or... I don't see a need for Monte Carlo here. — Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
ElMaestro ★★★ Denmark, 2009-11-27 21:01 (6053 d 01:56 ago) @ yes_naresh Posting: # 4396 Views: 4,336 |
|
|
Hi Naresh, could you explain what you are trying to achieve? Best regards EM. |
|
martin ★★ Austria, 2009-11-28 10:41 (6052 d 12:16 ago) @ yes_naresh Posting: # 4397 Views: 4,306 |
|
|
dear naresh ! you my find this textbook useful: Birkes D. and Dodge Y. (1993). Alternative Methods of Regression. Wiley, New York. hope this helps martin |
Ing. Helmut Schütz