Timeline for What's the point of time series analysis?
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 22, 2018 at 15:14 | comment | added | Lucas Roberts | +1. @Stephan Kolassa, your answer highlights how forecasts are used in the real world which is my interpretation of the OP's question. The prediction interval point and (half) length is exactly the information you care about for planning mitigation strategies. If you're building levee's to prevent Manhattan from flooding and your new time series method reduces the prediction interval by enough, you can reduce the cost of levee construction by using only the resources necessary. Mutatis mutandis applies for your grocery example. | |
| Sep 19, 2018 at 13:33 | comment | added | Stian | +1. It is useful elsewhere as well. Doing a time series analysis will certainly alert you to events (you did not know of) that affect a result you are interested in. I and all my colleagues were completely stunned to find we generally were sliiightly worse on product chemical analyses on tuesday mornings. We tracked it back to a well intentioned cleaning schedule that had some weaknesses. We saved over the year close to a million and improved the product ppk from 1.7 to 1.9. Lesson learned: always do a rudimentary time series analysis on any variation framed problem. | |
| Sep 19, 2018 at 8:51 | history | answered | Stephan Kolassa | CC BY-SA 4.0 |