Is sharpe ratio that good ... Is a trader with only 2%return,1%stddev as good as one with 20%return,10%stddev? The first one is just playing it safe.
I prefer performance index = %return X (1-%maxstdev) X no.of trades
Ratios can be ambiguous.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Help needed with probability math
- Thread starter braincell
- Start date
Is sharpe ratio that good ... Is a trader with only 2%return,1%stddev as good as one with 20%return,10%stddev? The first one is just playing it safe.
I prefer performance index = %return X (1-%maxstdev) X no.of trades
Ratios can be ambiguous.
You don't know the distributions from which these two samples were drawn.
Mean and standard deviation doesn't matter in this case because samples are very small.
A message to the usual suspect who insisted that average = expected value: You have no shame. Get a life.
Problem as posed is not solvable.
Mean and standard deviation doesn't matter in this case because samples are very small.
A message to the usual suspect who insisted that average = expected value: You have no shame. Get a life.
Problem as posed is not solvable.
Quote from intradaybill:
That is also what every crook with a ponzi scheme wants you to believe.
Well, okay, I shouldn't have phrased it as "The Answer", but I meant it in that it does answer the OP's question in the most succinct way -- i.e., how to compare the two systems with the same mean but one with broader variance. It is not an excuse for the small sample size. All statistics must be interpreted with caution, especially those based on tiny sample sizes. The best one can say about this data is that there is a "trend". There are no statistically significant conclusions one can make.
Quote from virtualmoney:
Is sharpe ratio that good ... Is a trader with only 2%return,1%stddev as good as one with 20%return,10%stddev? The first one is just playing it safe.
I prefer performance index = %return X (1-%maxstdev) X no.of trades
Ratios can be ambiguous.
One should not rely completely on a single statistic. It is what it is: a measure of trade-off between mean profit and spread. I personally would just look at the mean and std dev (profit and risk) separately -- e.g., one could plot mean vs std dev to decide on the optimal systems (upper left quadrant is good).
Oh christ in heaven... E[X*X] is the uncentered standard deviation, which is often used when the returns series offer unreliable estimates of the mean - there's life after stats101, you know...
Nevermind. OP - I'm done with the thread. Like so many threads before, I find myself underguned in patience and dignity compared to kut2k2. I suggest you might find better information over at the more learned forum catered to less 'elite' traders.
Nevermind. OP - I'm done with the thread. Like so many threads before, I find myself underguned in patience and dignity compared to kut2k2. I suggest you might find better information over at the more learned forum catered to less 'elite' traders.
Quote from kut2k2:
And notice the gyrations that top-posting mouthbreather DontMissTheShortBus has to twist himself into trying to "prove" that all I did was restate his Sharpe nonsense. Since when is E[x²] equal to the variance? Oh yeah, it's when E[x] is zero. But E[x] is emphatically not zero here. See how ridiculous these little math dilettantes are? Like undisciplined puppies soiling your carpet, they don't realize it unless you rub their noses in it.
Impossible to have a civil conversation with arrogant fools. My 2¢.