EssayStoryJournalBook reviewReflectionResearchLetterPoemMemoir462 words3 min readAutosave enabledMarket timing is the strategy of making buy or sell decisions of financial assets (often stocks) by attempting to predict future market price movements. The prediction may be based on an outlook of market or economic conditions resulting from technical or fundamental analysis. Attempts at market timing reflect the general desire to improve portfolio performance over a “buy-and-hold” portfolio. Yet, there is little evidence that market returns are predictable in the short term given that for efficient markets, there exists an equivalent martingale measure that ensure that the discounted stock return/price is a martingale. How then might it be possible to improve portfolio performance over a buy-and-hold portfolio? (don’t bother to understand what a martingale is for now..) Research has shown that it is possible to predict not the magnitude of future market returns, which is difficult if not impossible, but the volatility of future market returns, which are more quickly visible. Focusing on the volatility dimension may enable performance that is superior over a buy-and-hold portfolio. So one may ask, Is forecasting market volatility a form of timing? In a sense, yes, but it is a very different sort of timing and it can help inform the building of what we call “Volatility Managed Portfolios” (VMPs). Forecasting Volatility is Possible and Practical A large body of research documents the notable predictability of asset return volatility and the accuracy of volatility estimates as much higher than return estimates. For example, with 120 observations – 10 years of monthly data – the estimate for volatility is about 40% more accurate than the estimate for returns. Put another way, roughly speaking, it takes a year of returns to learn as much as we do from about two months of volatility calculations. An intriguing implication of forecasting volatility is that the future direction of the market, not the magnitude of the move, is also forecastable. When volatility is very low, the returns are likely to be clustered around the positive expected return and most realizations are positive. On the other hand, when the volatility is very high, there are more occurrences of negative returns, but also of very positive returns as well. An Example of Volatility Timing and Performance To see the implications of volatility timing for an investor, Tyler Muir and Alan Moreira (Muir and Moreira, 2015) compare a buy-and-hold portfolio with a portfolio that allocates capital to the market based on the variance of the market portfolio in the immediately preceding month. The dynamic portfolio adjusts so that, if the variance of the daily returns in the past month is higher, then the allocation to the market in the next month is reduced. As the graph below shows, this dynamic portfolio outperforms a buy-and-hold portfolio in a consistent manner from 1926 to 2015.