1
$\begingroup$

Among the several weighting schemes used for constructing volatility indices, which ones are the best for forecasting (realized) volatility? I've constructed a volatility index for emerging markets using a (very simple) trade-weighted approach, which has 84% correlation with the VIX (2007-2012). The construction procedure is much more simple, and I'm still running my regression analysis, so I can't say yet which one has better predicting power.

After reading some articles I've seen lots of critics to the VIX methodology. Is the VIX methodology the best one or there are competitive alternatives for weighting schemes?

TY

$\endgroup$
2
$\begingroup$

Just couple points to ponder about:

  • Weighting schemes are not a magic bullet and none of them in isolation will give you higher predictive power (aka. edge). Do not rely on how you weigh components.

  • Why would you want to be highly correlated with VIX? Its a poorly constructed index and I find the whole rational behind the implementation details quite questionable. VIX has a very low predictive power.

  • Markets are dynamic, most surely weights change, predictive power changes as everything else does: I encourage you to look a lot deeper than just how index components are weighted.

$\endgroup$

Your Answer

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