What If Stocks are Not for the Long Run?

Posted on June 2nd, 2009 by admin in Quant

We have discovered an important analysis of stock prices that is both obvious and surprising, not to mention useful.

Remember that old (legalistic) warning “past performance is no guarantee of future results”?  That’s obvious. But the implication is that investors believe the past is somehow useful to predict the future (lawyers aside, please). So here’s the catch: the view of prices looking backward is very different than looking forward. When you look forward, making decisions must account for substantially more risk than you can infer from pure historical statistics.

For example, the notion of “implied volatility” in options pricing is supposed to be the view of investors looking forward. But given the method by which this is computed (using only historical prices), it really is only a backward risk level. The view truly looking forward would give higher risk. Keep that in mind!

OK, we know many of you will be glazed like a donut by now, but in case you want to think deeply about the uncertainties of stock investing a good entry is this interview with the lead author ( the interview is actually quite understandable):

What if stocks are not for the long run?

(Read interview with Prof. Lubos Pastor, Booth School of Business, University of Chicago)

For you quants, here is the abstract of the published paper.

Lubos Pastor
University of Chicago – Booth School of Business; Centre for Economic Policy Research (CEPR); National Bureau of Economic Research (NBER)

Robert F. Stambaugh
University of Pennsylvania – The Wharton School; National Bureau of Economic Research (NBER)
May 22, 2009
Abstract:
Conventional wisdom views stocks as less volatile over long horizons than over short horizons due to mean reversion induced by return predictability. In contrast, we find stocks are substantially more volatile over long horizons from an investor’s perspective. This perspective recognizes that parameters are uncertain, even with two centuries of data, and that observable predictors imperfectly deliver the conditional expected return. Mean reversion contributes strongly to reducing long-horizon variance, but it is more than offset by various uncertainties faced by the investor, so that annualized 30-year variance is nearly 1.5 times the 1-year variance. The same uncertainties also make target-date funds undesirable to a class of investors who would otherwise find them appealing.

Keywords: stock, volatility, target-date funds, Bayesian, predictive system, predictive variance

JEL Classifications: G12

Working Paper Series

Leave a Reply

More News