Timeline for Forecast multiple regression in R
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 10, 2015 at 16:36 | comment | added | Richard Hardy |
I figured my comment qualifies as an aswer to the original question, so I posted it as such. Regarding your questions in the comments, the second one qualifies as a separate question. The first one should be "yes", I suppose. Try reading the documentation for the predict.lm function.
|
|
| Aug 10, 2015 at 16:25 | answer | added | Richard Hardy | timeline score: 1 | |
| Aug 10, 2015 at 14:31 | comment | added | Carol.Kar | @RichardHardy Thx for updating your answer! From my data atm, which variables would you choose for a multiple regression. It is hard for me to see this out of my data? Or: which tests should I be looking for to check for a lag? | |
| Aug 10, 2015 at 14:29 | comment | added | Carol.Kar | @RichardHardy Thx for your answer! Is it also possible to see the period of the forecast for each point forecasted value? Atm I am only get the values back.... | |
| Aug 10, 2015 at 14:25 | comment | added | Richard Hardy |
You can use the function predict instead, no calls for a new library are needed. But you need to supply new values of the independent variables to be able to predict the values of the dependent variable. If the dep. and indep. variables in the model are contemporaneous, you will not be able to forecast the future. You need a model where lagged values of indep. variables would match contemporaneous values of the dep. variable; then you can forecast the future.
|
|
| Aug 10, 2015 at 13:55 | history | asked | Carol.Kar | CC BY-SA 3.0 |