Back to Community
Institutional Level Long/Short Strategy (OCF/EV)

This is a strategy I developed in August. I wrote about the strategy and reviewed some of the pyfolio statistics here. It is extremely simple in that it relies on a single valuation factor - Operating Cash Flow / Enterprise Value. I would argue the strength of this algorithm is in the implementation. It attempts to be market neutral and sector neutral using a rebalancing corridor of +/- 5%. The target gross leverage is 2x. The strategy looks pretty solid but there are some weaknesses that I leave to the community to experiment with and/or test.

There is likely an unaccounted for, premium the strategy collects from trading less liquid shares. If you change the ADV filter to more liquid stocks the performance begins to degrade.

I tried to run the strategy using the new Q500/Q1500 portfolios and the backtest consistently times out so the evaluation with those standardized universes is incomplete.

I've only been able to review the strategy in the Research environment a couple times. Ideally I'd like to examine the in-depth trading statistics, but recently have not been able to without crashing the notebook. You may have better success.

As always the strategy could be subject to incorrect slippage and commission assumptions which could bias results up or down in real trading.

Additionally I do not believe Quantopian has incorporated the costs of shorting into the backtests therefore that could also present an upward bias in these results if those costs are significant.

Clone Algorithm
Backtest from to with initial capital
Total Returns
Max Drawdown
Benchmark Returns
Returns 1 Month 3 Month 6 Month 12 Month
Alpha 1 Month 3 Month 6 Month 12 Month
Beta 1 Month 3 Month 6 Month 12 Month
Sharpe 1 Month 3 Month 6 Month 12 Month
Sortino 1 Month 3 Month 6 Month 12 Month
Volatility 1 Month 3 Month 6 Month 12 Month
Max Drawdown 1 Month 3 Month 6 Month 12 Month
# Backtest ID: 582a133bac1120107aaeb41a
There was a runtime error.
2 responses

mind sharing the notebook?

Unfortunately I can't. As I hinted at prior, after the change to the research environment, these longer backtests consume all the memory available to me in before the backtest has finished loading. Feel free to clone the strategy and experiment with creating the notebook as you see fit.