LOGEST() results

  • Thread starter Thread starter Guest
  • Start date Start date
G

Guest

I was comparing results given between linest() and logest() for the same
group of data. For linest(), I had a std error of 7.02 and an r squared of
..6906, for logest(), I had a std error of 1.2689 and r squared of .346. Does
logest() report the numbers in another form? When I tested the real results
against the formula y = b* (m1^x1) *....*(mn^xn), the std error should be
higher.

Thanks
 
ReggieC -

As I recall, Excel uses a natural log transform of the Y values for fitting
its exponential function, implemented as the LOGEST worksheet function. So
it's likely the standard error is the standard deviation of the difference
between the log of actual Y and the log of fitted Y, i.e., quite different
from the linear case.

- Mike Middleton
http://www.DecisionToolworks.com
Decision Analysis Add-ins for Excel
 
Back
Top