1
ACCG329 Lecture 4: Informational
Efficiency and Model Evaluation
2009, Semester 1
Egon Kalotay
Overview of Lecture
• Finishing some topics from last week:
– Estimation of risk loadings using instrumental variables
– Testing and evaluating factor models
• We then consider Informational Efficiency:
– What does it mean?
– What does it imply?
• Theoretical
• Practical
pricing model evaluation
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
Some Observations from Lecture 3:
• Statistical assumptions give rise to a general pricing theory
that has a lot of appeal
• In principle, factor models provide a simple, flexible way
to approach the modeling of covariance risk and return
expectations
• Applications to asset allocation and various forms of
‘arbitrage’ trading
• Need to be careful in applications as there is little if any
consensus on model specification and the theory seems to
invite data mining
• Empirical evidence relating to out of sample usefulness is
best described as mixed
Single Index Model Estimation
Issues
• Beta estimates are more reliable at the
portfolio level
• The average single index model slope
coefficient (beta) is 1
• The more extreme a beta the larger the
likely estimation error
This is the end of the preview.
Sign up
to
access the rest of the document.
 Three '09
 EgonKalotay
 Economics, Trigraph, model evaluation, model estimation, th e th, Egon Kalotay

Click to edit the document details