Timeline for Would adding 0's to time slots that never have values improve the performance of any time series model?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 16, 2018 at 20:16 | answer | added | Chris Umphlett | timeline score: 0 | |
| Nov 16, 2018 at 19:23 | comment | added | John | Among things to consider: en.wikipedia.org/wiki/Zero-inflated_model | |
| Nov 16, 2018 at 17:41 | comment | added | John | Like I said, if you're dealing with something like sales per hour, then the data's not missing the way statisticians usually think about missing data. You observed zero sales. Beyond that, it's hard to say what is the optimal model to fit or anything else depending on what the data looks like and what you really are trying to do. | |
| Nov 16, 2018 at 17:35 | comment | added | robertwolfheart | For the sales per hour case, how would it make the model better? is there any explanation? | |
| Nov 16, 2018 at 16:41 | comment | added | John | It might be the right thing to do, but it depends on what you're modelling. For instance, suppose I am fitting a time series model to the S&P 500 index. Putting in zero will have a huge effect. Instead, if you're fitting a model to sales per hour, then zero is a reasonable value if missing. | |
| Nov 16, 2018 at 16:03 | history | asked | robertwolfheart | CC BY-SA 4.0 |