Defining Risk
It is a shame that much of the academic community in finance and portfolio management still uses standard deviation (and its cousin beta) as a measure for risk! It should be clear to everyone that standard deviation is no more than a statistical measure of historical volatility. Similarly beta is a measure of historical volatility relative to the market. Neither of them has anything to do with the forward risk of a security which is what investors really care about. In fact high volatility is not necessarily bad as long as the price goes up (or down if shorting).
For me, risk is simply the possibility of suffering a loss in the future. It depends on the purchase price and on whether the position is for long or short. If a stock or the market is undervalued, the risk ahead is very low for the long and very high for the short, regardless of its past volatility which is always the same for both the long and short at any given moment.
An example: the 12-month trailing standard deviation for the S&P 500 index in early 2003 was 3.6 times of what is now (see figure below). Does it mean that the market was much riskier at the time than it is now? Absolutely no! In fact early 2003 was the safest time to buy stocks and in that year the market turned out the highest return in recent memory. Though the volatility is quite low now (in fact it is the lowest in at least six years), the risk is probably much higher considering the state of the US economy and the level of share prices.
Some of my safest investments have been in stocks with high volatility and high beta (> 2.) at the time of purchase. They are the safest because all available data suggests that the stocks were way undervalued or that a new uptrend was about to start.
So why those smart academic people are still using standard deviation and beta? I can only think of two reasons: (1) these metrics are still somewhat useful for uninformed investors who fear volatility and for fund managers who cannot assess the true risk of stocks independently, hence float and sink with the whole market; and (2) without standard deviation or beta as a form of generalization, risk would be very company specific or time specific [which should be the case to start with], hence they cannot develop or teach students about any generalized theory or model, including the famous capital asset pricing model (CAPM) and, as a result, would loss their job.
[See Also:] My Investment Strategy
For me, risk is simply the possibility of suffering a loss in the future. It depends on the purchase price and on whether the position is for long or short. If a stock or the market is undervalued, the risk ahead is very low for the long and very high for the short, regardless of its past volatility which is always the same for both the long and short at any given moment.
An example: the 12-month trailing standard deviation for the S&P 500 index in early 2003 was 3.6 times of what is now (see figure below). Does it mean that the market was much riskier at the time than it is now? Absolutely no! In fact early 2003 was the safest time to buy stocks and in that year the market turned out the highest return in recent memory. Though the volatility is quite low now (in fact it is the lowest in at least six years), the risk is probably much higher considering the state of the US economy and the level of share prices.
Some of my safest investments have been in stocks with high volatility and high beta (> 2.) at the time of purchase. They are the safest because all available data suggests that the stocks were way undervalued or that a new uptrend was about to start.So why those smart academic people are still using standard deviation and beta? I can only think of two reasons: (1) these metrics are still somewhat useful for uninformed investors who fear volatility and for fund managers who cannot assess the true risk of stocks independently, hence float and sink with the whole market; and (2) without standard deviation or beta as a form of generalization, risk would be very company specific or time specific [which should be the case to start with], hence they cannot develop or teach students about any generalized theory or model, including the famous capital asset pricing model (CAPM) and, as a result, would loss their job.
[See Also:] My Investment Strategy
No comments:
Post a Comment