Free Site Registration

Maximizing Minimum Quantity: A Test for the Best

Traders Magazine Online News, March 27, 2018

Eric Stockland

 “What’s the best min quant?”

For years, brokers and exchanges have been fielding some variant of this question from the buyside. But while many practitioners have personal rules of thumb, it’s a tough question to answer empirically because few have the data to evaluate the effectiveness of different strategies. Until now.

Why use a minimum quantity?

So what is the objective of using minimum quantity? Firms have a variety of specific reasons for using a min quant, but it really all comes down to limiting exposure to intermediaries. And the data backs this up?—?our analysis shows that liquidity adders typically receive better markouts when the liquidity taker is not a proprietary trading firm.

Of course, some intermediaries play a useful role?—?bona fide market makers tighten spreads and can offer material price improvement?—?but they all have a finite risk limit and can only trade so much with a given investor or broker. Put simply, if you are a large buyer and buy from an intermediary often enough, that intermediary will eventually have to “cover its position,” making it a buyer like you. As a buyer, that intermediary will likely be more adept and potentially less patient than you, meaning they could impact the stock’s price and availability by buying before you’re finished buying. On the other hand, if, as a large, natural buyer, you transact with natural sellers who are not looking to cover, there shouldn’t be a ripple effect after the initial transaction.

The problem is that min quant does not allow you to directly filter out intermediaries, since naturals also trade in small sizes. So how can you avoid throwing out the proverbial baby with the bathwater?

Exchanges can help with data

Unfortunately, naturals can’t effectively experiment with min quants because they don’t know who they’re trading with or who they would have traded with if they didn’t use a min quant. Analyzing different min quant strategies requires raw data on orders and trades from both buyers and sellers, and the types of firms they’re from.

As a marketplace, IEX has the marketable order data and the trade data, and the firm types[1] for each, so we are in a unique position of being able to back-test min quant models to gauge how effectively they filter out intermediaries and small-sized natural flows.

How IEX tested min quant models

We tested two kinds of min quant models against a baseline of all marketable orders[2] or all trades,[3] using one month of IEX data. (August 2017)

  • 300 shares: A static minimum quantity of 300 shares
  • Dynamic models: 5%, 10%, 15%, 25%, 50% of the median NBBO size, set on a symbol by symbol basis and rounded down to the nearest round lot and with a minimum of 100 shares

For more information on related topics, visit the following channels:

Comments (0)

Add Your Comments:

You must be registered to post a comment.

Not Registered? Click here to register.

Already registered? Log in here.

Please note you must now log in with your email address and password.