Hi everyone,
Currently i am working on my thesis and i am stuck with the following question. Is it sensible to include control variables in my VAR model?
Right now i am using a VAR model to estimate the effect of Twitter sentiment on European index returns (E.g. DAX). I use both intraday (30/60min) and daily data to estimate the VAR model. However, i am not sure whether i need to include control variables into my VAR model. I was thinking about controlling for inflation, interest rate or GDP, but i have not found any other studies including these variables. Since i am a student i am only capable of finding daily data for the aforementioned control variable, so maybe it is more sensible to leave them out anyway.
Right now i am interested in estimating two VAR models:
1. Sentiment score on Index Returns
2. Tweet Volume on Index Returns
Is anybody able to help me? A link to an article or something else would also be very welcome.
Kind Regards,
Mark
Related Posts with Is it sensible to include control variables in my VAR model?
Pooled OLS versus random effects in an extremely unbalanced panel- Install Unix stata on Ubuntu in Windows Subsystem for Linux
- Two-way fixed effects and multiple FEs (some questions to clear things up with the models used)
- Weakly balanced panel
- How to compute for inverse Mills ratio from a heteroscedastic probit model
Subscribe to:
Post Comments (Atom)
0 Response to Is it sensible to include control variables in my VAR model?
Post a Comment