Computer-based models to replace the investment advisor?
By Bob McDowall, IE4C
Published Monday 21st August 2006 12:31 GMT
Analysis A number of recently published features and surveys evidence
the continued growth in quantitative investment.
Quantitative investment is based on the deployment of computer
generated investment decisions. It is reputed to be growing at 20 per
cent per year. Major conventional fund managers as well as hedge
funds are increasingly deploying computer-based models to make
investment decisions rather than relying on human judgement - or is
that a myth?
Quantitative investment managers use a model to identify sets of
characteristics for their investments. Computing power is now
relatively cheap. Obviously, computing power can access data almost
instantaneously and simultaneously. Asset classes and financial
instruments within those asset classes can then be screened and
investments are selected. They reflect the manager's views.
These models normally determine the investment decisions and so
replace the traditional portfolio manager's role. It is contended
that this approach eliminates human emotion and personal bias, which
can impede effective portfolio management.
More importantly, the models provide insight into market
inefficiencies to be applied rapidly across asset classes and the
vast number of financial instruments within those asset classes.
Whole markets can be analysed daily for buy and sell indications at
an individual instrument level. This enables portfolios to contain a
larger number of instruments and reduce risk through greater
diversification of the portfolio.
Computers have not taken over the investment process. Human qualities
are required to set the criteria and parameters for data collection
and analysis. Here the investment stars still have their place - a
creative and efficient human portfolio manager to extract the value
from quantitative data.
It is a complete myth that quantitative investment managers are
nerdish boffins working under the direction of an all-powerful
computer model. It is equally a myth, perpetuated by some technology
providers, that there is reliance on a "black-box" stereotype.
Dependency on programming and computer nerds has not replaced the
dependency on star investment managers.
Quantitative investment managers distinguish themselves by their
processes, "which combine people and technology in a framework that
rigorously assesses cost, risk and return, which sets them apart from
less successful managers".
Trading is a function, where it would be possible and highly
irresponsible to construct/build a completely automated quantitative
process. Trading is not a mechanical process. Trading is influenced
by relationships, trust (and the lack of it!) and gamesmanship.
Equally, trading relies on much more than the application of
intuition or "gut feeling". The key role of traders is to provide the
balance between pure judgement and systematic solutions.
For example, a principal programme trading, where a broker quotes a
price to guarantee completion for a list of trades and executes those
orders at current market prices, will encompass a model of expected
cost for assessing competing quotes from brokers. Yet, the traders'
selection of brokers best suited for each trade is critical to
successful execution.
Some quantitative fund managers let the data set the direction by,
for example, using computers to identify patterns and relationships
in asset classes and instrument prices. Others are sceptical of
investment ideas that originate in and from "black-box". By working
with theoretically sound investment themes, they should be
identified, rationalised and justified by people for performance to
be stable and continuous into the future
Quantitative portfolio construction is about taking forecasts of
asset class and individual instruments and the returns on them,
models of portfolio risk and estimates of transaction costs and,
finally, reconciliation of optimal trade-off between these
components. Importantly, it is accompanied by a high degree of
management and oversight.
The portfolio manager should vet the information going into the
creation of forecast returns. As with other forms of data entry
"rubbish in, rubbish out" is true. Portfolio managers, for example,
have check major changes in earnings forecasts and examine extreme
high and low valuations to make sure they truly reflect expected
earnings, and are not some form of data error. Longer-term economic
outlooks and less substantive issues, which may contribute to the
performance, have to be mapped against the data, as do extraordinary
events such as 9/11 or the recent terror alerts in the UK.
Quantitative management focuses on collection and analysis of
historical data. In consequence it is alleged to have a selection
tendency towards so-called "value-based" investment selections with a
strong bias to selection of companies with strong cash flow and solid
tangible assets. This ignores or places a lower rating on matters
such as quality of management, products and services, and innovation
not yet translated into cash flow such as intellectual property, for
example brands and patents.
Successful quantitative managers must be innovative, "seeking to
extract the best from man and machine" to produce a quantitative
investment process that is fast, accurate and exceptional in
execution. The process has to deliver consistent performance both of
markets and competitors in the asset classes with an ability to
adapt. In a fiercely competitive environment of financial markets,
this is a continual and evolving challenge. An active quantitative
investment approach must blend people's insight and creativity with
the efficiency and speed that technology can supply.
Investment performance is ultimately dependent on the quality,
innovation and insight of research. Quantitative managers should be
interested in all opportunities to outperform the benchmark return.
Ideas, which evolve and contribute to this achievement, will not come
from a machine but from people.
search: