Bernd - Thanks for updating this.

Just a point regarding objectivity of posts and web links. You're right
to point out limitations here, but i believe in letting users make up
their own minds based on individual merit without too much opinion.

In fact most formulas provided by posters will not return correct
results in all situations, instead they are balanced to the needs of
the given problem, too much error checking makes them unwieldy and
harder to understand. For example all formulas using match/lookup
functions for nonexact matches can return incorrect results if data is
not arranged appropriately.

I think most would agree extrapolation should be turned off by default
anyway, eg forecasting tomorrows FTSE based on yesterdays movement
is misguided, but that it can be occasionally useful.

Excel solutions are often designed around expediency and a RAD mentality.
The emphasis is usually on getting results - it's often impractical
to try and account for every conceivable turn of events along the way -
instead time is always spent on thoroughly checking the answer.
(this is how it works on financial services trading floors anyway.)



"Bernd P" wrote in message
...
Hello Lorimer,

Thanks. I included a link to that thread on my webpage.

To give your approach the right but limited credit:
http://www.sulprobil.com/html/poor_m...rpolation.html

IMHO all formulas should inform their user about their limitations
and, if they cannot provide a correct result, they should return with
an error message.

Regards,
Bernd