Dear All ,

I have a 100 sets of time series data (each consisting of 16 points) which
measure a certain feature of the company's performance.

We would like to test whether these time series can be categorized into
various groups using curve fitting, for example

Group 1 .... has a logarithmic form
Group 2 ... has a general cubic form
Group 3 .... has an exponential form

The current plan is to look at each time series, broadly classify them into
groups by eye, and then fit a curve to each member of each group using
nonlinear least squares.

This does not seem a very rigorous or sensible way of analysing these time
series. Does anybody know whether there is a systematic way of analysing
such a set of time series data to find common patterns between individual
time series ?

Thanks in advance

Paul Kerr-Delworth

Data Analyst
Kunick Leisure


"  Information and data sourced within Kunick PLC, its subsidiary and
associated companies ("Kunick") are of a proprietary nature unless
otherwise stated.Accordingly they must only be used in connection with
Kunick business.

The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential and/or privileged
material.  Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited.  If you receive
this in error, please contact the sender and delete the material from any
computer.

This footnote also confirms that this Email message has been scanned for
the presence of computer viruses.  "



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%