Liquidity provider algo idea

February 7, 2010

Here is the idea to utilize optimal trading strategy to extract excessive return through providing liquidity in the short term market. The basic tenet behind is that aggressive price taker demand immediate execution hence inccurs higher transaction cost. By providing liquidity to them and act to bridge the short term liquity and longer term liquidity, liquidity provider should be able to extract this part of premium impatient trader give up. using a pca factor model, the linear regression residue would be approximately reflect the idyosyncratic factor which will be dominated by erratic liquidity mismatch in the market, this could serve as an indicator to initiate market making algo.

p.s. I found this on internet: here (Doyne Farmer Andrew Lo 07)

Modeling Limit Order Removal Times from Market Data with q-Weibull Distribution <== this is the building block to estimate liquidity imbalance. actually working paper is missing(?)

Adlar J. Kim is student of Lo and Farmer at SFI, member of AI and Postdoc Sloan. very interesting combination. Adlar’s 08 thesis.


security trading of concept*?*

market making by learning liquidity imbalance ***

[]model stock order flow and mm * kim and shelton’s 02 paper, using learning in both market model and strategy model, rely on shelton’s o1 thsis.

hidden markov chain for model mm *

[]farmer’s presentation at SFI * basically double auction simulate mkt well.

Electronic market making * (CMU intelligent software agent Center here)

reenforcced learning for optimal trading * (Upenn micheal kearns another AI Market don as Farmer)

Lo’s auto MM patent and this with reference and related info worth digging

C.R. Shelton UC Riverside prof. machine learning, cowork with adler, micheal kearns et. al. publication list here. Shelton’s PhD thesis is on importance sample to improve POMDP (MDP with hidden states) with multiple objective, much close to some real world problem like MM.

an electronic market maker AI lab memo by nick chan and shelton.

Artificial market and intelligent agent, nich chan phd thesis.

here is an actual agent based forecast product, a very interesting idea Altreva.

Study of Artificial Financial Markets with Adaptive Trading Agents, nich chan & adler kim.

A learning market making in Gloston Milgrom model, by Sanmay Das (adler kim and Sanmay both from MIT learning center) Sanmay’s other publication and thesis.

O’hara’s review on market structure issue on market liquidity(BIS paper) O’hara’s publication

Columbia math course (Financial Price Analysis spring 2010) outline: with an interesting list of reference

Alex Chehkov, one who offer the above mentioned course, is a visiting AP from Systematic Alpha Management, with an interest estimating liquidity imbalance as well. alex’s drawdown paper; something else; supply chain finance.

trading and market making course, Craig Holden do research in microstructure as well.


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: