Timeline for Dealing with varying predictive horizon
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 2, 2023 at 20:16 | comment | added | Darren Cook | @Marco See updated answer. | |
| Oct 2, 2023 at 20:15 | history | edited | Darren Cook | CC BY-SA 4.0 |
Expand answer based on question in comment.
|
| Sep 27, 2023 at 8:04 | comment | added | Marco Ballerini | Thank you again @Darren Cook, I’m sorry for my late reply but I’m still thinking of the problem. So, after I’ve trained my two models, let’s say P1 predicts what the result will be and P2 predicts how long we have to wait the result, how can I use them to have a single number telling me the probability of having the outcome 1 within a month? | |
| Sep 14, 2023 at 14:18 | comment | added | Darren Cook | @MarcoBallerini I think it is fine - if they are correlated it means the data is correlated. E.g. the higher months gets the less chance of result being a 1. I did just update my answer to clarify that only X1,X2 are used in each model. You can't use months to help predict target as you don't know it in advance. | |
| Sep 14, 2023 at 14:15 | history | edited | Darren Cook | CC BY-SA 4.0 |
Clarify that only the X predictors are used as inputs to each model.
|
| Sep 14, 2023 at 14:00 | comment | added | Marco Ballerini | Thank you @Darren Cook, you're saying I need to train two models (e.g two logistic regressions). What if the two models are correlated? Is there a way to take into account this issue? | |
| Sep 12, 2023 at 16:56 | history | answered | Darren Cook | CC BY-SA 4.0 |