liveBook · Manning

Adverse selection

As explained in Experimentation for Engineers

A trading condition in which execution is more likely when subsequent market movement makes the trade unfavorable.

Adverse selection occurs when execution is more likely just before market movement makes the trade unfavorable. The resulting lost profit is the adverse selection cost. section 4.2

Increasing order size can raise potential profit but also worsen execution quality: liquidity may be available precisely when the trade is about to lose money. The practical objective is to balance missed opportunity against adverse selection. section 4.2

Order-size trade-off

Requesting $3,000 of stock may execute only $2,000 because sellers do not provide the remaining liquidity. A stylized simulator models adverse selection cost as asc = 0.001 × exp(2 × order_size), where order_size is measured in thousands of dollars, then subtracts it from threshold-dependent markout profit. section 4.2

How Experimentation for Engineers covers it

How deep it goes
Defines the execution condition and models its cost exponentially, but stops short of estimating dynamics from observed data.
Examples
Uses partial execution of a $3,000 stock order and a Python markout-profit simulator with exponential adverse-selection cost.
What you can do after
Evaluate order-size and threshold choices with a two-parameter markout-profit simulation.
How it connects
Connects liquidity, order size, trading threshold, markout profit, and response surface methodology.

This explainer was generated by Manning's knowledge graph from the text of Experimentation for Engineers. The section links point to the parts of the book it draws on.