Turnover and Volume Filters

This stock trading question comes from Colin H. who asks:

“In the two books of yours I’ve read you suggest the use of both turnover and volume filters.  Whilst I like the idea I don’t have an empirical method for determining parameter values. For example, a turnover filter set to an average of $500,000 over seven days against the Russell 3000 would not work so well for the ASX Small Ords, or would it? Is there a method for parameter sizing based on universe depth and width?”

The key ingredient is the ability to enter and exit the market without moving it. I can create some outstanding results if I lower volume and/or turnover right down, but in reality I’m not going to be able to realize those simulated results if I can’t actually get in or out of the market in real time.

I’m only kidding myself with the testing – and plenty of providers do it. If you want some kind of benchmark then I would suggest your own position size should not exceed 10% of the daily volume or turnover of a specific stock. For example, if you have $10,000 position sizes then turnover should be a minimum of $100,000 for that stock. This benchmark can be applied equally across any universe.

Another consideration is the length of your holding period. If you’re a longer term trend trader then a little extra slippage won’t detract too much from the bottom line, so you may be able to lower your volume/turnover levels. However, if you’re a shorter term swing trader, then slippage will become a larger factor of profitability, so you want to err more on the side of caution.