Transcript Richard Michaud PhD

Portfolio Monitoring*
Richard Michaud, David Esch, Robert Michaud
New Frontier Advisors
Boston, MA 02110
Presented to:
QWAFAFEW NYC
September 27, 2012
* Forthcoming: Michaud, Esch, Michaud, 2012. “Portfolio Monitoring in Theory and Practice,”
Journal Of Investment Management.
© 2007 Richard Michaud and Robert Michaud
© 2011 Richard Michaud and Robert Michaud
About New Frontier
• Institutional research and investment advisory firm
• Inventors and authors in investment technology
• Michaud and Michaud, Efficient Asset Management, 1998,
Harvard, 2008., 2nd Edition, Oxford
• NFA is unique:
•
•
•
•
Institutional investors who use our own software
Global software providers who manage money
Published authors in books and refereed journals
Four U.S. patents, two pending
22
© 2011 New Frontier Management Company, LLC
Current Portfolio Monitoring Ad Hoc
 Calendar rebalancing
 Monthly, quarterly, yearly, three years, every five minutes
 Asset weight hurdle ranges
 Drifted portfolio relative to neutral or optimal weights
 Ranges typically vary based on asset volatilities
 No theory to support practice
 Not portfolio based rules
 Often trading in noise or not trading when useful
3
17
© 2011 New Frontier Management Company, LLC
True Portfolio Monitoring
 A statistical similarity test:
 Is the current drifted or given candidate portfolio
statistically similar or different relative to optimal
 If statistically similar, don’t trade
 If statistically different, trade
 Presentation scope:
 Decision whether or not to trade
 How to trade or how much to trade is a separate issue
4
17
© 2011 New Frontier Management Company, LLC
Academic Portfolio Similarity Tests
 Shanken (1985), Jobson and Korkie (1985), Levy and Roll
(2010)
 Tests of CAPM
 Is “market” statistically mean-variance (MV) efficient
 Limitations of academic tests
 Analytical tests assume unconstrained MV optimization
 Hotellings T2 and other analytic methods
 Not useful for investment practice
 Practice requires linear inequality constraints
 Constraints part of defining test statistic
 See Markowitz (2005) why constraints essential
5
17
© 2011 New Frontier Management Company, LLC
First Constrained Portfolio Similarity Test
 Michaud (1998, Ch. 7)
 Portfolio distance function relative to Michaud frontier
 Uses patented resampling technology
 Computes need-to-trade probability
 Relative to thousands of simulated investment scenarios
 Technology used in NFA’s World Gold Council reports
6
17
© 2011 New Frontier Management Company, LLC
Resampling and the Michaud Frontier
67
© 2011 New Frontier Management Company, LLC
Statistical Portfolio Monitoring Illustrated
88
© 2011 New Frontier Management Company, LLC
What the Monitoring Rule Computes
 Associated simulated optimal portfolios provides a distance
scale for monitoring portfolios
 Portfolio distance function (one example)
 Relative variance function = (P – P*) (P – P*)
 A measure of distance in N-dimensional portfolio space
 Sort distance low to high distribution
 Defines probability scale from 0 to 99%
 Compute distance from current to optimal
 Defines probabilistically how far current from optimal
9
17
© 2011 New Frontier Management Company, LLC
What the Rule Means
 10% need-to-trade probability means
 Portfolio distance is 10% as far as others in distribution
 75% or more probability may indicate trading is
recommendable
 50% probability often a useful default value
 Balance between avoiding noise trading and being able to
detect true deviations from optimality.
10
17
© 2011 New Frontier Management Company, LLC
Using Portfolio Monitoring Rule
 Decide on level of probability for trading
 L = Probability level for trading
 Recommend trading if probability > L
L depends on many investment and client issues
 Investment Styles:
 High levels -- value managers?
 Lower levels -- growth managers?
 Client Preferences, investment horizon
 Specialized investment classes
 Way to monitor universe of managed accounts
 Portfolio monitoring automation
11
12
© 2011 New Frontier Management Company, LLC
Limitations of the Original Michaud (1998) Rule
12
© 2011 New Frontier Management Company, LLC
Limitations of Michaud (1998) Test
 Low statistical power
 Infrequently rejects no-need-to-trade null hypothesis
 Poor power at high end of frontier
13
17
© 2011 New Frontier Management Company, LLC
Meta-Resampling Solution
 Patented meta-resampling (Michaud and Michaud 2002, 2008)
 Associates resampled with Michaud efficient portfolios
 Each simulated “parent” MV efficient frontier spawns a
“child” resampled efficient frontier
 Associated child resampled efficient frontier portfolios used
to compute distance probability
 Greatly enhanced statistical power
 Nearly uniform power across frontier
14
17
© 2011 New Frontier Management Company, LLC
Michaud Frontier Associated Meta-Resampled Portfolios
12
estimated average return (%)
10
8
6
4
2
0
0
5
15
© 2011 New Frontier Management Company, LLC
10
15
standard deviation (%)
20
25
Highly Compute Intensive Process
 Use better computer technology
 Multi-core computers
 Network multi-core
 Cloud computing
16
17
© 2011 New Frontier Management Company, LLC
Still A Persistent Problem in Practice
 Need-to-trade probabilities often seemed too
low in actual practice
17
17
© 2011 New Frontier Management Company, LLC
The Common Information Issue
18
© 2011 New Frontier Management Company, LLC
The Common Information Issue
 Information in current portfolio often based on similar information in
new optimal
 Common information means two portfolios similar all things equal
 Need-to-trade probability necessarily small
 Test is no-trading-biased in presence of common information
 Michaud-Esch-Michaud conditional monitoring rule
 A new scale that includes common information
 Dramatically enhanced power for many practical applications
 Realistically sensitive to changes in current vs. optimal
 Three levels of resampling in general case
19
19
© 2011 New Frontier Management Company, LLC
Illustrating Conditional Monitoring Algorithm
 One year ago optimal portfolio P0
 X0= [x1,x2,…,x60] = defines original risk-return distribution
 New optimal portfolio P*
 Xnew = [x13,x2,…,x72] = defines new risk-return distribution
 48 months of common information: [x13,x2,…,x60]
 Compute meta-resampled portfolios (simplest case)
 Compute k = random draws = 12 from Xnew distribution
 Add to common 48 months: [x13,x2,…,x60] = sim distribution
 Compute meta-sim optimal and distance to P*
 Repeat above many times
 Sort and define distance distribution
 Compute P0 distance to optimal and percentile in distance
distribution (conditional need-to-trade probability C(k))
20
4
© 2011 New Frontier Management Company, LLC
Actual Case: Conditional Monitoring Rule
21
© 2011 New Frontier Management Company, LLC
Applications
 A measure of regime changes in markets
 Assume a long-term strategic optimal portfolio
 In drifted period
 Minimal market volatility – little need to trade
 High market volatility – likely need to trade
 Return distribution generalizations
 Simulations can be based on any distribution
 We generally use t-distribution
22
4
© 2011 New Frontier Management Company, LLC
Summary
 Portfolio monitoring an essential asset management function
 Prior methods ad hoc, academic methods invalid
 Patented first practical monitoring rule Michaud (1998)
 Limited statistical power
 Patented Meta-resampling rule Michaud and Michaud (2002)
 Enhanced statistical power across frontier
 Customizable to asset management processes
 Michaud-Esch-Michaud conditional monitoring algorithm
 Common information, increased statistical power
 Highly compute intensive procedures
 Just finance catching up to real statistics
23
28
© 2011 New Frontier Management Company, LLC
Extensions
 Potential for large-scale automatable portfolio monitoring
 A statistical context for general quadratic programming applications
 Process monitoring and multivariate regression in the context of
linear constraints and overlapping data
24
4
© 2011 New Frontier Management Company, LLC
Thank You
New
NewFrontier
FrontierAdvisors,
Advisors,LLC
LLC
Boston,MA
MA 02110
02110
NFABoston,
SAA
Portfolios
www.newfrontieradvisors.com
www.newfrontieradvisors.com
25
25
© 2011 New Frontier Management Company, LLC
Richard O. Michaud






President, Chief Investment Officer
Co-inventor (with Robert Michaud) of Michaud Resampled
Efficient Frontier™, three other patents, two pending
Author: Efficient Asset Management, 1998. Oxford
University Press, 2001, 2nd Edition 2008 (with Robert
Michaud)
Many academic and practitioner refereed journal articles
CFA Institute monograph on global asset management.
Prior positions include:
 Acadian Asset Management; Merrill Lynch
 Graham and Dodd winner for work on optimization
 Former Director and research director of the “Q” Group
 Advisory Board member, Journal Of Investment
Management
 Former Editorial Board member Financial Analysts
Journal, Journal Of Investment Management
26
© 2011 New Frontier Management Company, LLC