We are doing a time series regression project to identify the variables that impacts the bitcoin price in USD. We came up with the following explanatory variables:

1) GDPs of US, Canada, Europe, China, Slovenia, Singapore and Japan.

2) Aggregate computing/processing power: We don't know which index to use here.

3) Inflation rates of the previously mentioned economies

4) Activity on TOR browser (downloads for instance)

5) Number of vendors accepting bitcoin payments

6) financial sector's and IT sector's GDP for the aforementioned countries

7) Crime rate and cyber crime rate in each period

8) Dummy variable for major events. (for instance the shut down of the Silk Road online marketplace by the FBI)

9) Stock indices such as S&P 500, DOW 30, Nikkei 225, NASDAQ, NYSE, and ...etc.

10) Inflation rate of the previously mentioned economies

If multicollinearity exists between some of the independent variables, they can be combined into 1 variable via linear transformation.

What do you think about the explanatory variables suggested above? Can you suggest other explanatory variables that would increase the R squared, AIC, BIC, and other related quality measures?

  • Looks like you copy/pasted this from somewhere. Maybe you can clean it up a bit? "up vote" ?
    – Jannes
    Nov 30, 2015 at 16:29


Your Answer

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

Browse other questions tagged or ask your own question.