1
$\begingroup$

I'm required to use two time series models in my exam project. I want to use a stock price of an energy company, and then explain it first using ARIMA, and then adding other variables and using VAR. My problem is that I can't find variables which pass the Granger test of causality. I have tried prices of commodities, competitor stock prices, etc.

So now I have opted to use two of the same company's stocks, but just listed on different stock exchanges (same company, same currency - one stock trading in Germany, the other in Brussels)

Obviously, the stock on both markets is impacted by the same unobserved variables.

Is this a huge mistake, or can I proceed if I discuss and show that I am aware of omitted variable bias?

Improve this question
$\endgroup$
3
  • $\begingroup$ In which way do you think it could be a mistake? What variables do you think you are omitting? VAR models are often used for forecasting where omitted variable bias is of little concern, thus my question. $\endgroup$
    – Richard Hardy
    Commented Jan 12, 2021 at 7:39
  • $\begingroup$ Thank you for your reply,Richard Hardy. The company mainly produces electricity. So, of course, using daily electricity prices for the German market could be a much more interesting variable, but it is not possible to access this data for free. And there are probably other variables that I can't think of due to lack of experience. But thank you for clarification that omitted var. bias is not highly relevant in this case. I will proceed with what's there and keep that in mind! $\endgroup$
    – Rakamakafon
    Commented Jan 17, 2021 at 14:19
  • $\begingroup$ Please visit stats.stackexchange.com/help/merging-accounts to merge your accounts: that will enable you to continue commenting on and editing your posts. $\endgroup$
    – whuber ♦
    Commented Jan 17, 2021 at 16:33

1 Answer 1

Reset to default
0
$\begingroup$

Maybe it's a bit too late to your exam. I think it's natural that you can't find relationship for stocks, which have high noise to signal. My preliminary thought is that if you want to stick with stock prices, maybe do a grouping based on industry, in your case, electricity stocks, create an index tracking the average return of the industry, and then do the GC test on this new series instead. I think this way it reduces noise and is more likely to find relationship that you are looking for.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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