|
PureBytes Links
Trading Reference Links
|
Does anybody have a particularly good reference to the methods of Engle
and Granger? These are ARCH methods. Anyone want to comment on any
known successful implementations of these methods in trading?
Here are a couple of links from the Nobel Prize Internet Archive which
have a Book Store section and a Featured Internet Links section:
http://almaz.com/nobel/economics/2003a.html
http://almaz.com/nobel/economics/2003b.html
Below is a brief excerpt from the NYT.
Rod
+++++++++++++++++++++++++++++++++++++++++++++
An American and a Briton have won this year's Nobel Memorial Prize in
Economic Science for developing statistical methods that allow
researchers, policy makers and Wall Street traders to better analyze
stock prices and other long-running series of data.
Robert F. Engle and Clive W. J. Granger spent much of their career
working together at the University of California at San Diego, doing
their seminal work in the 1970's and 80's. Their research has enabled
others to study the relationship between variables in ways not possible
before. A more sophisticated understanding of those relationships — what
is cause and what is effect, for instance — has led to a richer
understanding of how the economy works and better forecasts.
Before the two published their work, economists often had to assume that
a variable was no more likely to move in one direction than the other,
even when the evidence — such as the rise in stocks over the last 200
years — showed otherwise. Statistical models simply offered no
alternative. In the late 1970's, however, Professor Granger developed
ways to analyze the relationship between two statistics that had both a
long-term trend and an element of randomness.
Professor Engle's primary work improved the understanding of volatility,
particularly in the stock market, and enabled economists to make more
accurate forecasts. Previously, researchers were often forced to assume
that volatility did not change over time.
|