7 Comments
User's avatar
Ben Harrington's avatar

Great piece! did you rebalance SP/Russell every year or just use the current baskets? (constituents change every year). If so, where do you find historical constituent data? i’ve had a tough time finding that in the past trying to test strategies on index constituents

Quantitativo's avatar

Thanks! We only trade the constituents as they existed at each point in time, so there's no lookahead into future index membership. For historical constituent data, we use Norgate Data. They provide full historical index compositions, so our backtests are free of survivorship and selection bias. Can't recommend them enough if you're doing any kind of strategy testing on index constituents. Cheers!

rvadata's avatar

Thank you!

german_research's avatar

Amazing article and a genius idea !

alphacap89's avatar

Hi. Regarding the short side, are you still trading this as a long-short style? If so, how much leverage are you using, and is the short side a mirror of the long side? I have listened to the Marsten Parker episode here https://bettersystemtrader.com/183-overcoming-broken-strategies-marsten-parker/ and he raised an interesting point that a lot of traders make the "mistake" of flipping the 200 day sma rule for the short side because it makes the equity curve looks nice in 2008 but eats into performance during normal years so its better to still keep the stock above 200 sma rule even for the short side.

Ram Goli's avatar

Guessing this doesn't include trading costs?

Speaking of dynamic sizing... and you mentioned Ed Thrope, wondering if Kelly Criterion would be a natural progression :)

David's avatar

Which backtesting platform are you using for this? And which account size do you see as a cap for trading such strategies on stocks in the Russell 3000?