Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
ProShares launches anti-retail ETFs

From the FT this morning:

"The third [ETF being launched] resembles a traditional hedge fund offering a “long-short” strategy, buying stocks of companies active in online retailing while betting against bricks-and-mortar chains. All three will track bespoke indices managed by ProShares."

https://www.ft.com/content/dea75884-65bc-11e7-9a66-93fb352ba1fe

It would be a fun exercise to match this strategy on Quantopian.

7 responses

What data source would you use to determine whether a stock is primarily brick and mortar vs online?

You could divide land and improvements by total revenue, think the value would be higher for brick and mortar retailers. They might be selling their stores and leasing them back though.

Here's their SEC filing for the fund. I'm not sure but I suspect they will need to publish the composition or methodology of the index. I suspect this will be straightforward to implement in Quantopian, at least approximately.

https://www.sec.gov/Archives/edgar/data/1174610/000119312517224197/d416133d485apos.htm#toc416133_4

@atiredmachine You could look at book to market ratio (the traditional value factor). I suspect the online retailers will be low value, and traditional retailers will be relatively good value.

Looks like book value yield and tangible book value per share both show up under valuation ratios. Tangible book value yield might be a useful metric.

This ETF has now been released as of a few weeks ago.

I recreated in in Quantopian to see how the strategy/index performs historically.

My conclusion after studying this ETF is that their criteria for index inclusion is quite poor. For example, they don't differentiate between healthy, growing brick and mortar retailers with a big moat and ones drowning in debt and closing stores.

The strategy isn't a good fit for the Q Fund since it's not sector-neutral. It's also very high volatility. Also the types of improvements to the strategy that I think would be necessary to make it a winning strategy are hard to implement with the Quantopian framework.

Here's how my simulated ETF compares so far to the actual ETF as the benchmark.