36-1 investment manager selection
1 introduction
-due diligence
2 a framework
for investment
manager seach
and selection
3 quantitative
elements of
manager
search and
selection
style
analysis
capture ratios
and drawdowns
in manager
evaluation
defining the
manager
universe
errors in
manager
selection
the investigation and analysis in support of an investment action decision or recommendation
involves both quantitative and qualitative analysis
underlying
assumption:
a consistent robust investment process will generate a similar return distribution relative to risk factors through time
assuming underlying dynamics of market not dramatically changed
three broad
components
the universe
a quantitative analysis -
manager's performance track record
a qualitative
analysis -
investment
process
investment due diligence -
evaluates manager's investment process -
people / philosophy
operational due diligence -
evaluates manager's infrastructure and firm -
process / procedures
objective - to reduce manager universe to a manageable size relative to resources and time available
determine universe and benchmark - the IPS and reason for manager search
approaches
third-party categorizaton
return-based style analysis
holding-based style analysis
manager experience
*a hybrid strategy combine elements of each approach recommended
hypothesis testing - the null hypothesis:the manager is not skillful
type 1 - rejecting the null hypothesis of no skill when correct
type 2 - not rejecting the null hypothesis when incorrect
qualitative
considerations
predisposed
create explicit costs //
type 2 create opportunity costs
staightforward to measure and often directly linked to compensation
more transparent to investors
a consisten pattern of type 2 error - highlight weaknesses in manager selection process
the objective - to avoid making decision based on short-term performance and to identify behavioral biases
performance
implications
the smaller the difference in sample size and distribution mean and wider dispersion of distributions - the smaller expected cost of type 1 or type 2 error
the extent to which markets are mean-reverting -
has a bearing on the cost of type 1 and type 2 errors
the best manager: deliver desired exposures adn suitable for investor's assumptions expectatiosns and biases
the starting point - manager's self-reported risk exposure
to be useful
meaningful
accurate
consistent
timely
returns-based
style analysis
RBSA
top-down approach - estimating portfolio sensitivities to security market indexes representing a range of distinct factors
not subject to window dressing
an imprecise tool / illquid - stale prices
→longer periods
shoule consistent with manager's philosophy and investment process /
shoule tracked over time to ascertain - style drift
holding-based
style analysis
HBSA
bottom-up approach - estimates risk exposures from actual securities held in portfolio at a point in time
computational effort increase
necessary transparency come wih a time lag
some factors may difficult to estimate
holdings-based style map:
giant / large / medium / small / micro //
deep value / core value / core / core growth / high growth
capture ratios help assess manager suitability relative to investor's IPS in relation to time horizon and risk tolerance
useful in evaluating consistency
important to understand whether strategy inherently convex or concvexity derived from manager's skill
drawdowns - stress-tests of investment process and indicate potentially flawed or inconsistently implemented processes, inadequate risk controls or operational issues
*to worry more about type 1
*to assigning a manager to a benchmark