Timeline for Decomposing Historical Data - Arimax versus linear regression
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 5, 2019 at 20:12 | comment | added | IrishStat | 4 SURE ! ARIMA structure is an admission of ignorance as the past never causes the future | |
| Mar 5, 2019 at 14:24 | comment | added | B_Miner | I found a paper I think you wrote on existing store sales and new stores.....for autobox. Seems like a similar exercise. Is it true to think that a modeler for a marketing mix exercise like I outlined, should try to reduce the magnitude of the ar and ma terms through explanatory variables (or even eliminate the need for Arima errors altogether) as much as possible? | |
| Mar 4, 2019 at 21:04 | comment | added | IrishStat | Yes ... in this way you are accomodating for the impact of the X's that you know and the DETERMINISTIC X's (I's) that you don't know and the STOCHASTIC X's (e's ) that you don't know. | |
| Mar 4, 2019 at 19:25 | comment | added | B_Miner | Your comment about "...in general: means that this approach is valid? | |
| Mar 4, 2019 at 18:54 | history | edited | IrishStat | CC BY-SA 4.0 |
added 16 characters in body
|
| Mar 4, 2019 at 18:48 | history | edited | IrishStat | CC BY-SA 4.0 |
added 114 characters in body
|
| Mar 4, 2019 at 17:53 | comment | added | B_Miner | I assume by transfer function you how the effect of marketing is decayed over time (e.g. adstock)? I would look at that, this is just a quick example to see if regression with arima errors would be OK. | |
| Mar 4, 2019 at 17:52 | comment | added | B_Miner | I just changed the code section, clearly I was blind in my initial thought that the prediction from the linear regression was larger than the overall fitted value. Does this now look like a valid approach? | |
| Mar 4, 2019 at 17:46 | history | answered | IrishStat | CC BY-SA 4.0 |