5
$\begingroup$

I'm reading Vuolteenaho(1999). In this article, the author investigates whether the variation in stock market valuation level is driven by expected future cash-flows or by expected returns. In part V.B and table 5, the author decomposes the variance of log book-to-market ratio into components using GMM. I know the variance decomposition comes from the equation (16) on page 10 of Vuolteenaho(1999). But I can't understand how to estimate this decomposition by generalized method of moments. Could anyone give some detailed information about this decomposition?

I know GMM is a method of parameter estimation, and I know how to decompose variance using VAR model (like Campbell & Shiller (1988)). I just can't understand how to use GMM to variance decomposion.

Improve this question
$\endgroup$
2
  • $\begingroup$ He doesn't provide much detail but likely this is just a fancy way of saying that he estimates a theoretical population covariance by its sample value in data. $\endgroup$
    – fes
    Commented Aug 26, 2021 at 20:05
  • $\begingroup$ @fesman I guess so too but I'm not sure about it. Alter all, sample (co)variance is moment estimation of population (co)variance. $\endgroup$
    – Gödel
    Commented Aug 27, 2021 at 1:06

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.