with all due respect to the top managers at cisco who are complaining about the "100 year event" or "the perfect financial storm", on 9/21 the S&P was down about 37% from the top in April of 2000. based on the weekly data for US equity (Shiller, Siegel) for the period of 1871-2001, so far this has been the 20 year storm, or there-abouts.a -20% correction has a historical mean return period of 4.5 years, a -35% correction has a mean return period of about 17 years, and a -50% correction has a mean return period of about 45 years.the following graph shows the probability of not experiencing a correction of -20%, -35%, -50% in US equities over various time intervals extending out to 35 years. conversely, the probability of experiencing a correction at those magnitudes, based on the historical data would be (1-Ps).a note of caution - we used weekly closes to construct the graph so intraweek volatility is not represented. tr
tr,This may be a dumb question, but was the last 50% correction back in the 1930s, or did the S&P manage this with the tough bear market of the 1970s?
This graph just considers something a correction from the highest peak to the lowest trough, and then if that range is equal to or greater than the target percentage?
we used weekly data and just looked at the distribution of future drawdown for each and every data point. then we considered the distribution of the events with time, with the time stamp being the drawdown date.in other words - the events are not mutually exclusive - september 1974 represents various levels of drawdown depending on when capital was invested. continuing on with that example, for capital invested in 1960, 1973-74 did not represent a drawdown, whereas for capital invested in 1964 it did.tr
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |