Backtesting setup of a quant pm
Hello.. I wonder what backtesting setup you use as a quant researcher or pm? Do you use ready made python libraries (backtester or zipline) or in-house apps?. Thanks Adrian
Trending Content - Hedge Fund Forum
Career Resources
interested as well
I have a quick vectorized Python backtester for prototyping which I have written myself. It is very fast but does not simulate TC very carefully.
When an idea shows promise in this simple setup I submit to an in-house sophisticated simulator which could take hours to complete, but provides a good estimation of costs and other execution details.
Glad to hear from you..Thank you for detailed reply .I have a few systems to test..I know there are solutions like zipline but prefer to code myself.. Is Max drawdown as important as the Sharpe ratio at your fund?
As an extra question) Also if I have a multi exit logic how would you recommend to vectorize it ? In Excel I would set up separate columns for each condition and carry their trigger times up to see which one was triggered first.. In other words how do you vectorize future outcomes?
I would not say Sharpe Ratio is an important metric by itself. Realizing Sharpe of 10 with 500K revenue does not move the needle. However, realizing 100M revenue with Sharpe 1.5 does. DD is obviously important. Netting is netting but your strategies may experience DDs at the same time, you never know future correlations. Max DD is not enough to understand risks however.
In my experience vectorizing anything in Python comes down to writing the right np.where(...) (or similar functions) query. So, the advice is to just create an array of "triggers" and then slice it when necessary.
What is the holding period and turnover of your strategy? I suppose this is for intraday strategies that require sophisticated execution. Otherwise simulation time of a few hour is pretty long.
Even if your strategy has daily holding period you still need to rebalance intraday and deal with microstructure.
Excepturi est dignissimos similique quidem nostrum. At quam deleniti occaecati eos corrupti debitis vel et. Ipsum reprehenderit ea iste corrupti ea voluptatem dolore dignissimos.
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...