Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 0 down vote favorite

Previously, I did a basic course on econometrics which was mostly theoretical in nature and covered all the basic techniques up to cointegration.

I have this project in which I have a financial time series, and I want to be able to do some time series forecasting. In the course that I took, I didn't cover any time series forecasting, but I know a bit about statistical tests: DF tests, t-tests, White test etc.

The thing is: when you are trying to do financial time series forecasting, how do you start?

  1. How does the whole process work? My background is more theoretical than applied, so I'm looking for some tips here. I've never used R before, but I'm prepared to put in the effort to learn it.

    I have all my data in an excel file, and can export to csv if required. How then do I approach the problem of...

  2. Choosing an appropriate model (I have heard of ARIMA models but never studied them yet. I have a bit of experience with ARMA models.) So how would I know what kind of model is suitable for the data I have? Is there any general model that one can start with and modify? Any other models I should consider as well? Any pointers on this?
  3. If there was only 1 reference book you could recommend, what would it be?
  4. Currently, the only software that I can use would be R, which packages do you recommend? (in order of usefulness)
share|improve this question
1  
What financial market data are you looking at? – user13253 Sep 2 '12 at 5:04
3  
See otexts.com/fpp for an introduction on forecasting using R. – Rob Hyndman Sep 2 '12 at 5:12
i'm looking at fx now; note - the recommended book does NOT have to be free! i'm willing to pay for a gd reference bk! – user13783 Sep 2 '12 at 6:29
Check out this: epress.lib.uts.edu.au/scholarly-works/bitstream/handle/2100/…. The guy did a phd in maths then a phd in FX forecasting. – user13253 Sep 2 '12 at 7:31
"Financial time series" is a huge field. Basics start at AR and ARIMA. State Space approaches (Kalman-filter-based) give a lot of flexibility but are more programming-like. Also GARCH, VAR, ... As @Shaniqueia says, knowing what particular kind of data would help. – Wayne Sep 2 '12 at 13:40
show 2 more comments

closed as not a real question by Andy W, gung, whuber♦ Sep 2 '12 at 16:17

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, see the FAQ.