Re: Curve Fitting to Exponential Data

mcdonnellb@xxxxxxxxx wrote in news:1164116294.546782.136070
@j44g2000cwa.googlegroups.com:


Given the general formula for a exponential:
y = ae^(bx)


There's your main problem!! This is NOT the right model for the data set
as you've described it. You say there is a non-zero "horizontal"
asymptote, and you haven't placed it in your model

Try y=A*e^(Bx)+ C

Even better, you might want to MEASURE your asymptotic value, subtract it a
priori from your data, and then estimate your original model. This follows
a "never estimate what you can simply measure" rule of thumb.

Off the cuff, this looks a little double exponentially to me, or

y=Ae^(Bx) + Ce^(Dx) + F

I can't see the page with the method you link to for some reason, but you
might want to be a tad careful because the distribution of the residuals
depends so highly on X, and this could impact any inferential statistics
and confidence intervals you might get out of your fit.


--
Scott
Reverse name to reply
.

Relevant Pages

  • Re: Shark to EMC
    ... I would also add that the reason it can be done is the ... TDMF will not move at the data set level to fully use larger model disks ... XRC is the only method ... I can imagine would handle mirroring as opposed to moving the data. ...
    (bit.listserv.ibm-main)
  • Re: Modify DSCB information
    ... It depends on what you want to change and the reason for the change. ... Depending on how the data set is allocated at the time, ... Products like the old STOPX37 would intercept allocation process and add ... Search the archives at http://bama.ua.edu/archives/ibm-main.html ...
    (bit.listserv.ibm-main)
  • Re: Modify DSCB information
    ... It depends on what you want to change and the reason for the change. ... this to prevent space abends or some other reason? ... Depending on how the data set is allocated at the time, ... Products like the old STOPX37 would intercept allocation process and add ...
    (bit.listserv.ibm-main)
  • Re: Remember the imminent public humiliation of Paul Smith?
    ... a detailed reason for that and point out precisely what was wrong with ... a statistically significant reduction in fatal and serious ... To my mind an 11% reduction not being statistically significant implies ... either a small data set or a very fuzzy correction method. ...
    (uk.rec.cycling)
  • Re: overcoming zero values in line graph
    ... There is no reason why you cannot have a 2nd data set that has the N/A ... Plot this second set and use the first for whatever you are ... > is still zero. ...
    (microsoft.public.excel.charting)