Skip to content

Volume Functions and Liquidity Filters

October 18, 2010

Usually when testing strategies, it becomes important to have a basic minimum liquidity at place.

A lot of people filter out their universe based on market caps, some on volumes and further still on average traded value. All of them can be fine. Some performing better than the others. But one deep problem remains, all of them throw out otherwise perfectly tradeable assets/price series. In other words false negatives.

Before adding liquidity filters, we must really have a perspective into what do we want.Or rather what are we looking for.

Are small caps a problem, because they dont have enough market cap? Or is the price series essential “jumpiness” an issue? For me, it has something to do with the quality of price discovery .Ideally, if a stock’s price is 10paise, I am perfectly okay as long as it behaves like or nearly like liquid blue chip stocks, like LT or NIFTY. No huge jumps, no one sided dominated days, i.e it puts shadows on both sides etc etc. Very often, we involve ourselves with filters, which throw away a lot of otherwise perfectly tradeable assets. We do want to trade a small cap, which is liquid enough so that it doesnt whipsaw unnecessarily throwing our systems performance haywire


Define Illiquidity



What about this?


I reckon, having a measure of the “jumpiness” can help. How much does an asset gap up or down on average, or how frequent. An indicator with that kind of metric can help. Add with it,how often does it put shadows (that is Low!=Open High!=Close or vice versa).In my opinion, if you can find a linear combination of these two, then you will almost be very close to that level, which just sets the bar for a tradeable asset

Volume Functions

In my backtestings, involving volumes or any function there of into making decisions can improve the performance of the system. Whether, in a counter-trend system, shorting at some particular new high,only when your basic conditions are met,along withan additional condition of  no significant surge in volume being noticed. Or the opposite for a trend following system.

It makes sense to not only use the four basic market generated variables, OHLC but also V.

I could extend it a bit further and stick my neck out, and say use OI as well, but OI has argely disappointed me on a systematic basis.

No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: