10
$\begingroup$

My question is if there are any good online materials for learning this. Something that introduces things well, especially ARMA models and the related math.

Edit: I'm looking for something of the high-end undergraduate level. Something like in Brockwell and Davis' Introduction to Time Series and Forecasting

|cite|improve this question
$\endgroup$
  • 1
    $\begingroup$ Can you be more specific regarding what aspects of time series you are seeking to learn? At any rate, you should formulate your question to encompass a broader context than "my midterms are coming up", as otherwise it's almost certain to be closed as too localized. This should also be CW I believe. Cheers. $\endgroup$ – cardinal Mar 9 '12 at 18:11
  • $\begingroup$ I'm not really sure how to be specific without being too localized at the same time. $\endgroup$ – Mr Alpha Mar 9 '12 at 20:08
  • $\begingroup$ How much experience do you have with statistcs or math? Are you looking for implementations in a specific software (R? SAS?)? Are you looking for the theory behind ARIMA? You let us know what you are looking to do and I am pretty sure someone can help you out. $\endgroup$ – asjohnson Mar 9 '12 at 21:46
  • $\begingroup$ Both the Duke andthe Zoonek site offer very introductory material. Unfortunately real world data contains unusual values that require Pulses,Level Shifts,Seasonal Pulses,Time Trends to be treated within one model.Additionally parameters maychange over time, variance of the errors may change over time or in concert with the expected value or simply as an ARIMA model itself. These considerations require more computational fire power than is available for free or with most commercial offerings.A good way to learn time series is by actually trying to model or studying how an expert has modeled it. $\endgroup$ – IrishStat Mar 10 '12 at 20:50
  • 1
    $\begingroup$ For additional introductory material, the online statsoft textbook is often a good place to start, but I suspect @IrishStat is right in the end. $\endgroup$ – gung♦ Mar 10 '12 at 22:11
6
$\begingroup$

As per my comment I am not certain what you are looking for, but when I am fitting time series after a bit of a hiatus from them I tend to grab my copy of Time Series Analysis and Its Applications for more theory questions and I look at a few different sites online (also do some googling to see if there are any sweet new ones):

The CRAN taskview on time series gives you a good look at just how many things you can do

Is a nice walk through of some time series analysis in R. I personally do much of my statistical learning through example (which generally means following guides like this in R), so this guide is a favorite of mine.

This link is a decent look at ARIMA outside of R, it walks you through what different models mean.

Finally, you can always check out wikipedia if you are just looking for statements for formulas. These are just the ones I have book marked, so maybe some other folks will contribute their favorites. As I said in comments, if you expand on what you are looking for more specifically you can probably get better links from me or one of the folks that follow time series closer than I do.

|cite|improve this answer
$\endgroup$
  • $\begingroup$ Thanks for your answer, I had been trying to find that Duke website for ages, but had forgotten the university name and so wasn't getting very far! I wish I could give you more than +1. $\endgroup$ – Michelle Mar 9 '12 at 22:49
4
$\begingroup$

Personally, I think this webpage is a fantastic resource for time series analysis, particularly since it provides R code.

|cite|improve this answer
$\endgroup$
1
$\begingroup$

There are some good, free, online resources:

  1. The Little Book of R for Time Series, by Avril Coghlan (also available in print, reasonably cheap) - I haven't read through this all, but it looks like it's well written, has some good examples, and starts basically from scratch (ie. easy to get into).
  2. Chapter 15, Statistics with R, by Vincent Zoonekynd - Decent intro, but probably slightly more advanced. I find that there's too much (poorly commented) code, and not enough explanation thereof.
|cite|improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.