Timeline for Plotting predicted values in ARIMA time series in R
Current License: CC BY-SA 3.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 5, 2021 at 15:35 | comment | added | IrishStat | The following paper provides the reasoning/logic underlying AUTOBOX's feature which can distinguish between deterministic trend and stochastic trend. The former employs level shift indicators and time trend predictor series series along with possible pulses and seasonal pulses AND arima structure.  | |
| Apr 5, 2021 at 15:25 | answer | added | Caleb Watts | timeline score: 0 | |
| Jun 29, 2018 at 10:56 | vote | accept | Antoni Parellada | ||
| Jun 27, 2018 at 9:15 | answer | added | mbt | timeline score: 1 | |
| Nov 6, 2017 at 13:27 | comment | added | Antoni Parellada | @StephanKolassa I haven't gone back to explore time series in quite some time, and reviewing the OP, I have the impression I pretty much stuffed in there all the "tricks" I could come up with. I understand your comment (and Glen_b's) regarding non-linear trends. | |
| Nov 6, 2017 at 12:52 | comment | added | Stephan Kolassa | Antoni, have you found what you were looking for, and could you perhaps even self-answer? In any case, I second @Glen_b's comment: if you have a constant trend, then the differencing in ARIMA makes sense; but if the trend changes over time, then something like a state space model or double exponential smoothing looks better. Global temperatures are a case of the latter, as are stock markets on a daily or lower scale. | |
| Sep 29, 2017 at 14:07 | history | tweeted | twitter.com/StackStats/status/913767278625349632 | ||
| Aug 16, 2016 at 18:29 | comment | added | Antoni Parellada | Thank you. I am making some progress. It's a lot of fun! | |
| Aug 16, 2016 at 18:24 | comment | added | Richard Hardy | Rob J. Hyndman's blog post "Constants and ARIMA models in R" is probably all you need to know. I would be curious to hear you opinion once you explore the blog post. | |
| Aug 16, 2016 at 1:54 | comment | added | Glen_b | Rob Hyndman's comments here are relevant. I may come back and expand on that a little. | |
| Aug 16, 2016 at 0:25 | comment | added | Antoni Parellada | Thank you, @Glen_b. Just trying to get a flair for time series, and as in many math topics the lack of motivating preamble is a killer. All time series that we may really care about seem to trend up or down - populations, GOP, stock market, global temperatures. And I get that you want to get rid of the trends (may be for a second) to see cyclic and seasonal patterns. But the splicing back of the findings with the overarching trend to make predictions is either implied or not addressed as an objective. | |
| Aug 16, 2016 at 0:15 | comment | added | Glen_b | ctd ... There are a few other authors that do reasonably well, but even the better ones make it a bit more complicated than it really needs to be for a beginner. | |
| Aug 16, 2016 at 0:14 | comment | added | Glen_b | I would say that if you think you have a series where the trend has changed over time, ARIMA models may not be the best way to approach prediction of them. In the absence of subject matter knowledge (which might lead to better models), I'd be inclined to look at state space models; in particular variants of the Basic Structural Model for something like this. Many discussions of state space models can be hard to follow, but Andrew Harvey's books and papers are quite readable (the book Forecasting, Structural Time Series Models and the Kalman Filter is pretty good, for example). ... ctd | |
| Aug 15, 2016 at 17:27 | history | edited | Antoni Parellada | CC BY-SA 3.0 |
added 301 characters in body
|
| Aug 15, 2016 at 16:43 | history | asked | Antoni Parellada | CC BY-SA 3.0 |
