tag:blogger.com,1999:blog-35364652.post1671574524459140010..comments2017-06-25T08:54:36.837-04:00Comments on Quantitative Trading: In praise of day-tradingErnie Chanhttp://www.blogger.com/profile/02747099358519893177noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-35364652.post-51946579370076299462008-12-04T08:15:00.000-05:002008-12-04T08:15:00.000-05:00Hi Russ,You have an interesting point, but I am no...Hi Russ,<BR/>You have an interesting point, but I am not sure that the error estimate on the Sharpe ratio is solely determined by the inverse of the Sharpe ratio. It should be inversely proportional to the square root of the number of observations required to form that Sharpe ratio as well. (Clearly, if we just pick 3 consecutive winning days to calculate the Sharpe ratio, it will be very high indeed, but not really indicative of its long-term value!) <BR/><BR/>Hence I think both you and I are correct: the error on the Sharpe ratio is inversely proportional to both the Sharpe ratio and the sqrt of the number of observations. And for a fixed observation/backtest period, the number of observations is of course proportional to the frequency of trading.<BR/><BR/>Thank you for your insight.<BR/>ErnieErnie Chanhttp://www.blogger.com/profile/02747099358519893177noreply@blogger.comtag:blogger.com,1999:blog-35364652.post-42763694800611273542008-12-03T18:56:00.000-05:002008-12-03T18:56:00.000-05:00Actually how long it takes to verify (or disprove)...Actually how long it takes to verify (or disprove) a trading system has nothing to do with how frequently it trades. Fundamentally the question has to do with the error estimate on the Sharpe ratio, which is inversly proportional to the Sharpe ratio itself. Thus a very high Sharpe ratio strategy can be verified in a relatively short period of time. Low Sharpe ratio strategies take forever (10yrs or more) to achieve any kind of statistical significance. <BR/><BR/>A particularly obvious (and unrealistic) example is that of a strategy that never loses. All you need is a single loss to disprove it:)<BR/><BR/>I'd agree that in general it is the high frequency strategies that have high Sharpe ratios. And that therefore they are more easily confirmed than the others.Russnoreply@blogger.comtag:blogger.com,1999:blog-35364652.post-89936931926930208422007-09-08T21:08:00.000-04:002007-09-08T21:08:00.000-04:00Day Trading get's it's bad rap for good reason. Th...Day Trading get's it's bad rap for good reason. <BR/><BR/>The old day trading model has beed displaced with the new - but nobody got the word out. <BR/><BR/>John<BR/>www.DayTradersWin.compeppyhttp://www.blogger.com/profile/03943026837143119317noreply@blogger.comtag:blogger.com,1999:blog-35364652.post-60330945684397601412007-02-10T18:54:00.000-05:002007-02-10T18:54:00.000-05:00I believe the problem is that even when day-trader...I believe the problem is that even when day-traders have their emotions in check, and know what they’re doing with respect to entries and exits, all too often they do not also have a statistically viable system whose expectancy is greater than the cost of trading. Costs, alone, can kill you, and beyond that, if your position size or variance is too large, well..., let’s just say it’s easy to blow yourself up. Day trading is best left to those with the necessary skill-sets and experience and who are also not swayed by advertisements that tend to lure those who are most unsuspecting. The problem is as much the industry’s fault as it is anything else.Jayhttp://www.blogger.com/profile/02895545450530160682noreply@blogger.com