Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Assessing the Asymmetric Information Associated with the Equity Market A CART Based Decision Rule Analysis
1. Assessing the Asymmetric Information
Associated with the Equity Market:
A CART Based Decision Rule Analysis
Owen P. Hall, Jr., P.E., Ph.D.
Pepperdine University
CART Conference
May, 2012
San Diego, CA
3. Problem Statement
Assess the effectiveness of analytics to
detect asymmetric information associated
with the equity market
Models
• Probabilistic Neural nets
• CART
Factors
• Classic (e.g., Price Momentum)
• Tobin’s Q
• Entropy
4. Challenge
In an efficient market, the current
prices of securities represent
unbiased estimates of their true or
fair market value at all times
This principle suggests that neither
technical analysis nor fundamental
analysis can assist investors in
identifying undervalued or
overvalued stocks
I'd be a bum in the street with a tin cup if the markets were efficient
-- Warren Buffett
6. Entropy
The basic idea is that more volatile
securities have a greater entropy state than
more stable securities
Two fundamentally different phenomena
exist in which time based securities data
deviate from constancy:
Exhibit larger standard deviations
Appear highly irregular
The standard deviation measures the extent
of deviation from centrality while entropy
delineating the extent of irregularity or
7. Entropy
Two entropy models
Approximate entropy (ApEn)
Sample entropy (SaEn)
Model inputs
Time series
Matching template length (M)
Matching tolerance level (r)
Time series length (50 months)
8. Tobin’s Q
Q = Market value / Replacement value
Reflects the expected current and future
profitability of capital
Q values less than one identify under
valued equities
Q values greater than one suggest than
capex will increase share holder wealth
Q values less than one suggest making
acquisitions is cheaper than capex
11. Database
2008 (4) – 2010(1) – 6 Quarters
Sources
Value Line Investment Survey
Ford Equity Research
Mergent Online
Sample Size (100 ~ 400)
Target Variable – PGQ (binary- lagged)
12. Two Step Analytic Process
Screen variables with
neutral nets
Develop decision rules
using CART
Holdout Assessment
13. Probabilistic Neural Networks
An extension to the classical backward
propagation neural net
Non-parametric
“Black Box”
Results often difficult to interpret and
operationalize
15. CART
Non-parametric
Interactive effects
Non-normally distributed variables
Decision tree logic makes it easier to
apply model outcomes
Model is extremely robust to the effect of
outliers
Results easy to interpret and implement
17. Neural Net Results
Rank 8-4 9-1 9-2 9-3 9-4 10-1
1 PSS ROA PSS SMO CNE PRM
2 PRM SUE PVA PSS EMO Q
3 PVA PSS SEP ROA SMO ROA
4 ROA SMO PRM EMO VMO VMO
5 SEP EMO SMO SEP PEG EMO
6 VMO SEP Q Q Q SMO
7 EMO PRM VMO PRM SUE PSS
8 SUE PVA PEG EMO PRM PER
9 PEG PEG EMO PEG PSS SEP
10 SMO VMO ROA PVA SEP COM
18. Classification Analysis
(9/4 -> 10/1)
Actual
Predicted 1 0
1 31 15 67% PPV1
0 16 33 67% NPV2
Total 47 48
66% 69%
Sensitivity Specificity
PPV = ratio of the number of winners classified correctly divided by the total number of securities classified as winners.
1
NPV = ratio of the number of losers classified correctly divided by the total number of securities classified as losers.
2
20. Conclusions
Modeling approach generally
performed as well or better than
Valueline 100
CART results provide an
operational strategy
Adding transaction costs reduces
model effectiveness
Portfolio size based on binary
target variable remains
problematical
21. Future Research
Expand data set from 6 to 12
quarters
Ternary classification target
Variable selection optimization
Add economic factors
CPI
UEM
Explore “super” factors
Q / ApEn
PRM / SpEn