I am interested in analyzing disparities in volatility in different five minute intervals in different trading days in the Forex market. Now I want to know whether the volatility in such intervals can be grouped into clusters and whether the clusters are the same for different trading days. Can anyone help me by suggesting a method or technique to achieve my objective of identifying clusters for volatility?
Hi Hasanthi! I think you can try normal variance tests (Chi sq. and F-test) with moving or non-moving variances. In the first case, you can test whether the variance across the whole time series is constant. In the other one, compare two vairances, whether they are different or not.
Dear Hashanthi, first of all you have to carry out a heteroskedasticity test, to see whether your series can be modelled by GARCH model. There are many test, the most well-known is of course the Engle's test based on the Lagrange multiplier (Engle, 1982). Once you have identified heteroskedasticity, you can apply variance tests on different time intervals.
Hi. If you assume your processes are markovian, I suggest this nonparametric method to estimate trend and volatility of the processes and clusterize them:
e Gregorio, Alessandro & Maria Iacus, Stefano, 2010. "Clustering of discretely observed diffusion processes," Computational Statistics & Data Analysis, Elsevier, vol. 54(2), pages 598-606, February. The method is implemented in the R package yuima: https://r-forge.r-project.org/projects/yuima/ .
Why not taking the squared returns (or absolute returns) over the five minutes, take averages over days (for each five minute interval) and test the equality of means?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics