| Corporations
with large amounts of data often need to determine the degree
to which different datasets are correlated. We have developed
an automated correlation engine that was used to measure the
correlations between stock prices and economic indicators
such as interest rates, inflation indices, and foreign exchange
rates.
The correlations were calculated on a daily basis based on
well established daily volatility measures used in industry.
The resulting datastream was then visualised using SomFin
(see Visualisation).
|
|
Does the price of eggs
have anything to do with the rate at which people open bank
accounts or buy time share? If you have sufficient data
we can provide you with a number between -1 and 1, where
the value 0 means "no, the one has no relationship
with the other", the value -1 means "negative
correlation", i.e. if egg prices go up, time share
sales go the other way, and 1 means perfect positive correlation,
i.e. they move up and down together .
This tool may prove useful
for investigating whether sales patterns of certain products
are correlated in any way - product sales that are negatively
correlated will allow for the retailer to "hedge"
himself should sales in one of the lines decrease, by stocking
more of the other.
|
