Variance Ratio Test

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

randerson026
Posts: 1
Joined: Tue Jan 30, 2018 7:50 am

Variance Ratio Test

Postby randerson026 » Tue Jan 30, 2018 12:01 pm

Hello all,
I working on testing whether vol is mean-reverting using a variance ratio test. For fx_vol, I find the VR test statistics are small but I can't reject the null that the series is a random walk. Can I say that it is a random walk and potentially mean reverts (this may be a stupid question).
Thank you, Robin Anderson

Null Hypothesis: FXVOL is a martingale
Date: 01/30/18 Time: 12:54
Sample: 1/02/1990 1/31/2018
Included observations: 6426 (after adjustments)
Heteroskedasticity robust standard error estimates
User-specified lags: 20 40 60 90 120 260 520 780

Joint Tests Value df Probability
Max |z| (at period 20)* 2.294017 6426 0.1616

Individual Tests
Period Var. Ratio Std. Error z-Statistic Probability
20 0.568760 0.187984 -2.294017 0.0218
40 0.547880 0.251183 -1.799960 0.0719
60 0.552061 0.290595 -1.541453 0.1232
90 0.539930 0.330574 -1.391727 0.1640
120 0.491145 0.358723 -1.418519 0.1560
260 0.359012 0.430831 -1.487794 0.1368
520 0.276576 0.502254 -1.440356 0.1498
780 0.215461 0.552954 -1.418815 0.1560

*Probability approximation using studentized maximum modulus with
parameter value 8 and infinite degrees of freedom

Test Details (Mean = -0.00176937441643)

Period Variance Var. Ratio Obs.
1 0.09651 -- 6426
20 0.05489 0.56876 6419
40 0.05288 0.54788 6393
60 0.05328 0.55206 6368
90 0.05211 0.53993 6343
120 0.04740 0.49114 6329
260 0.03465 0.35901 6250
520 0.02669 0.27658 5971
780 0.02079 0.21546 5704

Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 17 guests