Timeline for How to forecast multivariate time-series 'accurately' with a large number of unknown factors using R?
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 5, 2014 at 19:02 | comment | added | forecaster | also, see the excellent answer by @Irishstat for this questions. stats.stackexchange.com/questions/66825/…. Looks like you are trying to solve a similar problem. | |
| Jul 5, 2014 at 18:44 | comment | added | forecaster | before taking the time series route,I would simply build a naive model and see how it predicts. I would also use structured judgement to adjust for special events etc., | |
| Jul 5, 2014 at 15:56 | answer | added | drenerbas | timeline score: 2 | |
| Jun 9, 2014 at 10:38 | comment | added | IrishStat | @Tim Please feel free to contact me at [email protected] and we can have a dialogue . | |
| Jun 9, 2014 at 3:27 | comment | added | Tim | @IrishStat I am still working on this problem at the moment but am just not making any headway. What further information would you need to help point me in the right direction methodology wise? and software wise? Given the time-frames this is really now just a personal project because I intuitively know it should be possible and am annoyed that I haven't been able to work it out. | |
| May 22, 2014 at 4:41 | comment | added | Tim | Regarding the 'diagnostic plots' I used to determine whether a forecast is 'good' or 'bad'; I don't know what the standard is for this but I looked at the percentage difference between the forecasted values and the actual values for each hourband and determined that the interquartile range was ~50-150% with a mean of ~70%. This level of inaccuracy is unacceptable for the purposes that I want to use the results for. | |
| May 21, 2014 at 23:04 | comment | added | Tim | Ok that's fair enough. What I have asked for is basically free professional advice and I agree that in principal that is abusing the forum. What I probably should have included is that this project is more about me personally trying to advance my knowledge than it is about what my employer wants, hence the 'no budget' part. I am allowed to use the data but its not a 'sanctioned' project. From my perspective, if i can get this to work then I gain a bunch of useful skills/knowledge and as a by product my employer gets a useful tool. I'm not sure where that leaves this question ethnically. | |
| May 21, 2014 at 13:08 | comment | added | IrishStat | I concur with @rocinante comments. It appears to me that there are two aspects to your problem/opportunity. The first (which I am quite capable of handling/consulting on) is methodological in nature i.e. what specific procedures are needed and the second question is what software is appropriate/cost effective. | |
| May 21, 2014 at 7:53 | comment | added | rocinante | As a "rule" this forum is not for giving out free statistical consultations to businesses who are unable/unwilling to pay professionals. Moreover, your question is extremely difficult to answer without seeing the diagnostic plots, what you used to determine whether a forecast is "good" or "bad". | |
| May 21, 2014 at 4:44 | history | asked | Tim | CC BY-SA 3.0 |