[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[RT] Correlation of international markets?



PureBytes Links

Trading Reference Links

JP Morgan has a "Risk metrics Group" that offers a free value at risk
calculator (online only). This text is from the website, it describes
some of what is available: 

"RiskMetricsŪ is the world's most popular portfolio model for evaluating
market risk. The RiskMetrics methodology enables a risk manager to
calculate Value-at-Risk (VaR) on a portfolio of financial instruments. "

"a methodology describing the RiskMetrics model; its inputs, assumptions
and limitations. The methodology is presented in the RiskMetrics
Technical document, available for download from the publications section
of this site in 'pdf' format. Updates to this document are available in
the form of the RiskMetrics Monitor."

"data sets that are freely available from this site. The data sets
contain consistently calculated volatilities and correlation forecasts
for use in estimating market risks. The asset classes covered are
government bonds, money markets, swaps, foreign exchange and equity
indices (where applicable) for 31 currencies, and commodities. Data sets
are provided for a 1 day and 1 month horizon."

http://www.riskmetrics.com/index.cgi

http://www.riskmetrics.com/rm/cde/index_varcalc.cgi

 "Use this page to select subsets of the RiskMetrics datasets to
Calculate Portfolio VaR. This service is free. Select assets from the
categories in the above frame. For each asset class, choose whether you
want All or a Selection of items. If you choose Selection, you will be
prompted to make further choices.When you have completed your selections
click on the 'Portfolio Values' market value of items in your portfolio.
When you are ready, click on 'Calculate VaR'. The Total VaR of your
portfolio as well as Incremental and Marginal VaR of each instrument will
be calculated. If you wish to load a named list of selected items and
parameters that you have previously saved, click on 'User Profiles'
above."
 

I think this may be more useful then simply looking at a correlation
matrix.

Ron McEwan