What is considered a successful quantitative trading strategy
Right now I am sophomore and I started making some quantitative trading strategy from ideas I have had. I was hoping someone with more experience could tell me what deems a strategy successful. Lately I have been looking at percent return, holding period, successful (positive yield)/ unsuccessful (negative yield) trade ratio, and trades per a year to determine if my strategy is good. Although, I do not know which characteristics are the most important. If someone could rank them in most to least important that would help me a lot. For example one of my strategies has an average return per a trade of 1.09% with about 40 trades successful to around 15 trades unsuccessful over a span of about 10 years with an average holding period of about 3 days. If I make the strategy more aggressive it will have an average return of .4% with about 650 positive yield trades and about 475 negative yield trades with an average holding period of about a day over the same 10 year span. As a result I am not sure which strategy is considered better. Obviously the total yield for the 10 years is higher when the strategy is aggressive; although, there is a higher success rate and yield per a trade when the strategy is more conservative.
Sharpe ratio should be the most obvious starting point for you, but only a starting point, mind you...
Very little experience, but I have dabbled a bit with backing testing strategies in matlab and python, and read a book on developing strategies. So I guess that makes me an expert haha
Obviously sharpe ratio is the big one because it incorporates the volatility of a strategy as well as the overall return. Drawdown is another popular metric, which is the maximum loss over the course of a series of sequential trades. Some traders can't stomach a high drawdown number, even if the sharpe ratio and overall returns looks promising.
Make sure you read about some of the biases for backtesting strategies. Data snooping bias is over fitting a strategy to historical data. Yes, it might have a good sharpe ratio, but that doesnt indicate furture returns will be good too. You need to find a way to randomly test your strategy in order to avoid this bias, so therefore you strategy knows how to handle "black swan" events (events that have never occured before in the stock market).
Other meaningful quick filter measures you could look at are the Sortino Ratio, MAR (=CAGR/MaxDD.). Schwager Gain-to-Pain and R^2 of equity curve.
Quia facilis architecto quia consequatur ea occaecati molestiae libero. Vel repellendus fugiat consequuntur neque.
Et voluptatem omnis modi. Saepe odio consequatur sit dignissimos qui animi. Voluptas eaque commodi ea eaque sed asperiores est harum.
Alias sed qui non omnis odio et. Quia ut asperiores veritatis. Aut quos nisi assumenda in et nemo quia sunt.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...