Market Microstructure Tutorial R/Finance 2009

Dale W.R. Rosenthal1

Department of Finance and International Center for Futures and Derivatives University of Illinois at Chicago

24 April 2009

UIC ICFD
[email protected]
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Who Am I

Assistant Professor of Finance at UIC Teach Investments; Market Microstructure; Commodities. B.S., Electrical Engineering, Cornell U., 1995. Ph.D., Statistics, U. Chicago, 2008. Programmer Intern, Listed Equities, Goldman Sachs, 19934. Strategist, Equity Derivatives, LTCM, 19952000. Trader/Researcher, Equity Trading Lab, Morgan Stanley, 2000-3. Self-employed Trader/Researcher/Coder, Summer 2004.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

What Microstructure Is and Is Not

Market microstructure studies the details of how markets work. Market microstructure is not neoclassical nance2 . often directly opposes Ecient Market Hypothesis. If you believe markets are ecient: the details of how markets work are irrelevant since. . . you always get ecient price (less universally-known fee). Thus microstructure embraces: the possibility of short-term alpha; and behavioral eects.

UIC ICFD
So you short one stock... omits a lot.
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

What I Cover in a Microstructure Course

When I teach microstructure, I cover many topics:
1 2 3 4 5 6 7 8 9 10 11

Market Types Orders and Quotes Trades and Traders Market Structures Roll and Sequential Trade Models Prices, Sizes, and Times Liquidity and Transactions Costs Market Metrics Across Time Electronic Markets Electronic Trading Tools and Strategies
Dale W.R. Rosenthal

Play with some

Strategic Trader and Inventory Models of these today.

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

Introduction

Today well consider models for microstructure phenomena. For these models, need to adopt a dierent perspective. Models are simple; lets us conduct controlled experiments.
Eliminate all but one or two major factors. Question: Does this model reproduce real-life features? If so: factors we are considering probably matter. We may even have an idea about how those factors matter.

Less confusion about what matters helps build better models. Newest research combining such models with time series. All models are wrong; some models are useful. George Box

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Asymmetric Information Models

We will examine two asymmetric information models.

Asymmetry: Trades may contain private information.

Sequential trade: independent sequence of traders.

Traders: informed (know asset value) or not; trade once. Single market maker; learns information by trading.

Strategic trader: one informed trader can trade many times.

Informed trader considers own impact on later trades. Uninformed (noise) traders also submit orders. Single MM sees combined order, sets price, lls order.

Some slides are for you to study later; Ill skip those.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom (1985) Model

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten and Milgrom (1985) Model

Most famous sequential trade model. MM quotes a bid B and ask A. Security has value V = V or V , V < V . At time t = 0, informed traders (only) learn V . Time is discretized. Trades are for one unit, occur at each time step. MM has innite capital: no inventory/bankruptcy concerns.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Setup

V V

w .p. 1 w .p. Informed Uninformed buy sell w .p. . w .p. 1

Trader type T

Traders take action S (buy/sell from MM), one at a time. Informed traders: S if V = V . if V = V

Uninformed traders: S

buy w .p. 1/2 . sell w .p. 1/2

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Event Trees

1 j5 buy jjjj / sell 3 I ; S =? 0 ggggg V = V ; T =? WW 1/2 kkk5 WWW+ kkk 1 / sell U; S =? 1/2

{{ {{ V =? CC CC CC C! C

1 { { {

{= V

t=0

0 3 I ; S =? SS / buy ggggg SSS) g 1 V = V ; T =?W sell WWWW+ 1/2 1 U; S =? T / buy TTTT )

t = 1, 2, . . .

1/2

sell

V = security value (V , V) T = trader type (Informed/Uninformed) S = traders side (buy/sell)

Dale W.R. Rosenthal

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Likelihood of Buys and Sells

What is P(buy), P(sell) after one trade? P(buy) = P(buy|V )P(V ) + P(buy|V )P(V ) (1 ) + (1 + )(1 ) 1 + (1 2) = = 2 2 P(sell) = P(sell|V )P(V ) + P(sell|V )P(V ) (1 + ) + (1 )(1 ) 1 (1 2) = = 2 2 (1) (2) (3) (4)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Likelihood of V , V

After one trade, we have information via Bayes Theorem: P(buy|V )P(V ) (1 )(1 + ) = P(buy) 1 + (1 2) P(buy|V )P(V ) (1 ) P(V |buy) = = P(buy) 1 + (1 2) P(V |buy) = P(sell|V )P(V ) (1 )(1 ) = P(sell) 1 (1 2) P(sell|V )P(V ) (1 + ) P(V |sell) = = P(sell) 1 (1 2) P(V |sell) = (5) (6) (7) (8)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Bid, Ask, Spread

If competition narrows prot to 0, before trading. . . A = E (V |buy) and B = E (V |sell). A = V P(V |buy) + V P(V |buy) (1 ) (1 )(1 + ) =V +V 1 + (1 2) 1 + (1 2) B = V P(V |sell) + V P(V |sell) (1 + ) (1 )(1 ) +V =V 1 (1 2) 1 (1 2) AB = 4(1 )(V V ) 1 2 (1 2)2 (9) (10) (11) (12) (13)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Updating Bids and Asks

After each trade, Bayesian update of beliefs about V . Idea: How often would we see these trades if V = V vs. V ? With these ideas, we can update the bid and ask prices. Expected bid and ask after k buys and Ak+ = = V P(V |k + 1 buys, sells) + V P(V |k + 1 buys, sells) V (1 + ) + V (1 )(1 + ) k+1 (1 ) + (1 )k+1 (1 + ) (1 )(1 + ) Bk+ = = = V P(V |k buys, + 1 sells) + V P(V |k buys, + 1 sells) = V (1 )k (1 + ) +1 + V (1 )(1 )k (1 + ) +1 (1 )(1 + )k (1 ) +1 + (1 )k (1 + ) +1
Dale W.R. Rosenthal

sells:

(14) (15)

)k+1 (1

)k+1 (1

(16) (17)

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Simulation Example

One simulation for = 0.3, = 0.5; example bids and asks. Simple case: V = V = 2 (versus V = 1). Can see price impact especially for sequence of orders.
2.0 Price 1.6 1.8

SBSBBBBSBSBSSBBSBSSBBSSBBBBBBBSBSSBBBBSBSBBSBBBSBB

1.4 0

10

20 Time

30

40

50

UIC ICFD

Dale W.R. Rosenthal

Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Simulation Averages

Simulate to nd the average bid and ask. 10,000 simulations; tried = 0.3, 0.5, = 0.3, 0.5. Simple case: V = V = 2 (versus V = 1).
2.0 2.0 Price 0 10 20 Time 2.0 2.0 30 40 50 1.4 0 1.6 1.8

= 0.5

Price

1.4

1.6

1.8

10

20 Time

30

40

50

= 0.3

1.8

Price

Price

1.6

1.4

1.4 0

1.6

1.8

10

20 Time

30

40

50

10

20 Time

30

40

50

= 0.3
Dale W.R. Rosenthal

= 0.5

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

Glosten-Milgrom Model: Other Results

Basic idea: Spreads exist due to adverse selection. Buys/sells are unbalanced; but, price series is a martingale. Orders are serially correlated: buys tend to follow buys.
This and the preceding line seem contradictory. Dierence is akin to Pearson versus Kendall .

Trades have price impact: a buy increases B and A. Spreads tend to decline over time as MMs gure out V . Bid-ask may be such that market eectively shuts down3 If uninformed were price sensitive, spreads would be wider. Code in THE SECRET DIRECTORY (glosten-milgrom.r). Fun: Add very rare third trader, govt, who always buys at V .
Stunning: still converging after 50,000 trades.
3

UIC ICFD
We nd ourselves in the game-theoretic Paradox of Trade.
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Kyle (1985) Model

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle (1985) Model

Kyle (1985) proposed a model with a single informed trader. The informed trader:
Considers price impact in setting trade size; Learns securitys terminal value v ; and, Submits order for quantity x.

Liquidity (noise) traders submit net order u. The single market maker (MM):
Observes total order y = x + u; Makes up the dierence; and, Sets the market clearing price p.

All trades happen at one price; no bid-ask spread. All trading occurs in one period.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: Informed Trader vs. MM

The informed trader would like to trade aggressively. Q1: What sort of function maps v to order size? A: Linear function? (Yes.) MM knows larger net orders are more likely to be informed. Thus MM sets price increasing in net order size. Q2: What sort of function maps order size to MMs price? A: Linear function? (Yes.)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: Setup

Security value Noise order MM assumes: informed order Net order Informed assumes: trade price Informed trader prot

v N(p0 , 0 ) u
2 N(0, u )

(18) u v (19) (20) (21)

x = v + y =x +u p=
illiquidity

y +

(22) (23) (24)

= (v p)x = (v (x + u) )x

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: Optimize, Compute Statistics

If we combine the previous formul, we get a little further: E () = E ((v (x + u) )x) = (v x )x
E (u)=0;x non-random

(25)

v if > 0 2 xR 1 , = = 2 2 p0 E (y ) = + E (v ) = + 2 2 Var(y ) = Var(x) + Var(u) = 2 Var(v ) + Var(u) x = argmax E () = = 0 +

2 2 u

(26) (27) (28) (29) (30) (31)

Cov(y , v ) = Cov( + v , v ) = 0

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: Find Linear Parameters

Now solve for the linear MM pricing and trader order parameters. MM earns no expected prot4 , prices trade at p = E (v |y ). Since v , y normal, form of E (v |y ) is like linear regression. p = E (v |y ) = E (v ) + Cov(v , y ) (y E (y )) Var(y ) 0 = p0 + 2 (y p0 ) u + 2 0 = p0 ; = (32) (33)

Use (33), (27), and (22) to solve for , , , : u = p0 ; 0 u = ; 0 0 . 2u (34)

UIC ICFD
Due to competition, again.
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: MM Price; Informed Order, Prot

0 y + p0 2u 0 2 0 = Value uncertainty Var(v |y ) = 2 u 2 u + 0 2 (v p0 )u Informed order x = v + = 0 (v p0 )2 u Expected prot E () = (v x )x = 2 0 MM trade price p = E (v |y ) = y + =

(35) (36) (37) (38)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: Commentary

What can we learn from the one-period Kyle model? Trade price linear in net order size, security volatility. Trade price inverse to noise order volatility. Informed order linear in securitys deviation from mean. Expected prot quadratic in securitys deviation from mean.
Large deviations matter much more than small deviations5 .

Informed order, expected prot linear in noise order volatility6 . Var(v |y ) =

0 2

Half of information7 leaks after one trade.

Negative net order (i.e. u < x) yields negative price. (!)

5 6

Except data errors and adverse selection will hurt you. Use uninformed orders to hide. 7 Information in a Fisher sense.
Dale W.R. Rosenthal

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

One-Period Kyle Model: Illiquidity Parameter

Finally: Consider the illiquidity parameter.

Illiquidity parameter = 2u0 . o /u is ratio of volatilities.

Value uncertainty vs. noise order uncertainty.

y =

y 0 2u : like liquidity risk:

Scaled by volatility of security; and, y /u is similar/proportional to percentage of volume.

Nice: Demanding liquidity has a cost. Full course covers more such ideas.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle (1985) Model

Kyle also discussed a multi-period model. Slice time t {0, 1} into N bins, n = 1, . . . , N: Time: Noise order: Informed order: Price change: E(Later prot): tn = 1/N, tn = n/N. un
2 N(0, u tn ).

(39) (40) (41) (42)

2

xn = n (v pn1 )/N. pn = n (xn + un ). E (n |pt<n ) = n1 (v pn1 ) + n1

(43)

Competition pn = E (v |x1 + u1 , . . . , xn + un ) Informed orders not autocorrelated, by construction.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Further Denitions

Use the following denitions: Strategy: X = (x1 , . . . , xN ). Pricing rule: P = (p1 , . . . , pN ). (44) (45)

X chosen to always maximize expected future prot: E (n (X , P)|v , p1 , . . . , pn1 ) E (n (X , P )|v , p1 , . . . , pn1 ) n = 1, . . . , N. (46)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Dynamics Equations

Model dynamics are given by dierence equations for n = 1, . . . , N:
2 n1 = n + n 2 u tn ; n

(47) (48) (49) (50) (51)

n1 = 1/(4n (1 n n )); 1 2n n ; n tn = 2n (1 n n )
2 n = n n /u ;

and,

n = (1 n n tn )n1 . n is the middle root of the cubic equation: 1 2 (1 2 u tn /n )(1 n n ) = . n 2

(52)

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Solving for Dynamics

Solving for the model dynamics is a bit crusty:
1

Guess N (call the guess ). N Use N = N = 0 to get N = Set n = N. Solve for n and . n1 Find n1 for n .

2 3 4 5 6 7 8 9

N . 2tN

Solve (52)8 , using middle root for n1 . n = n 1; if n > 0, go to step 4. If | 0 | > : try another , go to step 1. 0 N Solve for 0

UIC ICFD
Numerically or via Cardanos formula.
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Simulation

We can simulate the Kyle model to see its behavior. Use p0 = 2, 0 = 0.4, and u = 0.5. Run one simulation. What do we get? The parameter evolution is not so surprising. The action evolution is more illuminating.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Evolution of Parameters I

delta 0.08 0.00 0 0.04 0.12

10

20 Time

30

40

50

alpha

0.0 0

0.1

0.2

0.3

0.4

10

20 Time

30

40

50

beta

10

15

20

25

10

20 Time

30

40

50

UIC ICFD

Dale W.R. Rosenthal

Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Evolution of Parameters II9

lambda

1.05 0

1.10

1.15

1.20

10

20 Time

30

40

50

Sigma

0.0 0

0.1

0.2

0.3

0.4

10

20 Time

30

40

50

UIC ICFD
E.B.
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Evolution of Actions

u i i i i i i i i u i i i i i i i i i i i i i i i i i i i i i i i i i iui i i i iui i i i i i i iui i i u u u u u uu u u u u u u u u u u u u u u u uu u u u uu u uu uu u u u u u u u u u u u 0 10 20 Time True Price 2.4 30 40 50 0.010 Orders Trade Prices 2.0 2.1 2.2 2.3 -0.010 0.000

10

20 Time

30

40

50

UIC ICFD

Dale W.R. Rosenthal

Market Microstructure Tutorial R/Finance 2009

Multi-Period Kyle Model: Commentary

A few details nobody has previously noted10 : Informed orders11 are larger after negative uninformed trades. Informed orders decrease after larger net orders, price moves.
Recall: one leads to the other; confounding lives here.

Informed order size increases slightly with time. Trade price moves toward the true value. Trade price may not converge to true value by end of trading. Code in THE SECRET DIRECTORY (kyle.r).

10 As with the Glosten-Milgrom model, I have yet to see plots from anybody else who has simulated the Kyle model. 11 Informed trades are shown as is; uninformed trades as us. Dale W.R. Rosenthal

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

If You Want More: to Read

Journals (and associated societies):
Journal of Financial Markets Journal of Business and Economic Statistics/ASA Journal of Financial Econometrics/SoFiE Journal of Financial Economics Journal of Financial and Quantitative Analysis Review of Financial Studies/SFS

Books:
OHara, Market Microstructure Theory Harris, Trading and Exchanges Hasbrouck, Empirical Market Microstructure Weisberg, Applied Regression Analysis Montgomery, Design and Analysis of Experiments McCullagh and Nelder, Generalized Linear Models Box, Jenkins, Reinsel, Time Series Analysis Osborne and Rubinstein, A Course in Game Theory
Dale W.R. Rosenthal

UIC ICFD

Market Microstructure Tutorial R/Finance 2009

If You Want More: to Interact With

Seminars (times may change next year)

UIC Finance MSCS Fri Wed 10:30 4:15 Northwestern Finance IEMS Stat Wed Tue Wed 11:00 4:00 11:00 E&S Thu 1:20 U. Chicago Stat FinMath Mon Fri 4:00 4:30

Center Events: UIC ICFD, NWU Zell, UofC Stevanovich Conferences: ASSA, SoFiE, Oxford-Man Institute

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009

If You Want: to Support Work Like This

Talk to academics at the conference; some glad to consult. Take courses through UIC External Ed:
Market Microstructure and Electronic Trading Commodities, Energy, and Related Markets Fixed Income/Structured Products Empirical Methods for Finance Univariate and Mutivariate Time Series Analysis

Donate to the ICFD.

UIC ICFD
Dale W.R. Rosenthal Market Microstructure Tutorial R/Finance 2009