Timeline for Problem in discrete valued time series forecasting
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 25, 2012 at 14:54 | history | bounty ended | CommunityBot | ||
| Mar 20, 2012 at 15:16 | vote | accept | nb1 | ||
| Mar 19, 2012 at 18:11 | vote | accept | nb1 | ||
| Mar 19, 2012 at 18:11 | |||||
| Mar 18, 2012 at 18:47 | comment | added | nb1 | @ConjugatePrior hmm, I see. I will try out the packages once more as told by you. thanks. | |
| Mar 18, 2012 at 14:58 | comment | added | IrishStat | @ConjugatePrior you are quite correct I did assume some form of ordinal data. Now if it looks like a duck and quacks like a duck then it might be a duck . It is interesting to both of us how a "random selection" of ordered "labels" could exhibit evidently ordered structure BUT it can happen randomly and might just have happened here. | |
| Mar 18, 2012 at 11:37 | comment | added | conjugateprior | Mostly from the visuals, and because an AR 2 model fits reasonably, which I would not expect from an arbitrary set of numerical labels, though without a permutation check that's not more than curious. And perhaps a little because @IrishStat just flat assumed they were. | |
| Mar 17, 2012 at 22:41 | comment | added | nb1 | @ConjugatePrior yes, those are category labels. May I know why are you suspicious about the data being categorical? | |
| Mar 17, 2012 at 17:24 | comment | added | conjugateprior | Before thinking about that, I'd like to register a little, umm, concern about the problem setup. Are 1-6 really category labels? It seems a bit suspicious (statistically speaking) that such a neat autocorrelation story fits them. It seems more likely that we are dealing with a continuous process that generates ordinal data: a different (though not totally different) problem... | |
| Mar 17, 2012 at 16:38 | comment | added | IrishStat | This is quite interesting to me . Can you please detail your suggested approach with the OP's data. Please show as much detail as you can along with all of your assumptions and the actual forecasting equation. Please provide the arithmetic to show how one actually forecasts out the next 14 periods. | |
| Mar 17, 2012 at 16:30 | history | answered | conjugateprior | CC BY-SA 3.0 |