I created a time series for Terrorism data to see its effect on stock market's log returns using daily observations. I used cointegration technique for long run relation, after confirmation for long relation between the two studied variables, I applied OLS technique and noted that terror and log returns are positively related, which was different from earlier studies. Then I went back to my cointegration test result and noted that normalized cointegration vector are showing negative relation between the two studied variables. Now two different test are giving two different results.
After doing some literature review, I probed that same data set using wavelet technique and found different result for different time scale. Which make some sense but now I seek to get robustness of that result.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA