[SciPy-User] [Numpy-discussion] Fitting a curve on a log-normal distributed data
Gökhan Sever
gokhansever@gmail....
Tue Nov 17 16:21:27 CST 2009
On Tue, Nov 17, 2009 at 1:40 PM, Robert Kern <robert.kern@gmail.com> wrote:
> On Tue, Nov 17, 2009 at 13:36, Gökhan Sever <gokhansever@gmail.com> wrote:
> > On Tue, Nov 17, 2009 at 12:57 PM, Robert Kern <robert.kern@gmail.com>
> wrote:
>
> >> So, I would say that it's a bit trickier than fitting the log-normal
> >> PDF to the data for a couple of reasons.
> >>
> >> 1) Directly fitting PDFs to histogram values is usually not a great
> >> idea to begin with.
> >> 2) We don't know how much probability mass is in the censored region.
> >
> > So we agree that it is easy to implement a log-normal fit than a discrete
> > one?
>
> No, none of the things we have suggested will work well for you. You
> have a more complicated task ahead of you. I have ideas that might
> work, but explaining them will take more time than I have.
>
Looking at some recent replies and re-reading them a couple times, I should
say the techniques mentioned in them are beyond my technical skills or at
least I need a professor to help me or a good statistics book to study
further. I should also note that this is just a feasibility study comparing
actual observed cloud condensation nuclei concentration measurements to the
modelled concentrations using another instrument's size distribution data
with the help of a thermodynamic particle activation equation which I will
be able to infer an activation size limit. The results that are found in
this study will not be placed on a journal, they will just be presented in
my cloud physics class presentation. I am trying to assess the sources of
errors and testing the usability of the size distributions from that
particular instrument in this comparison study. Extending the size
distribution beyond and below the instruments measurement limit is one of
the biggest source of errors to represent the reality, but of course there
other simplifications and assumptions that add uncertainties.
Besides, what is wrong with using the spline interpolation technique? It
fits nicely on my sample data. See the resulting image here:
http://img197.imageshack.us/img197/9638/sizeconcsplinefit.png (Green line
represents the fit spline)
>
> --
> Robert Kern
>
> "I have come to believe that the whole world is an enigma, a harmless
> enigma that is made terrible by our own mad attempt to interpret it as
> though it had an underlying truth."
> -- Umberto Eco
> _______________________________________________
> SciPy-User mailing list
> SciPy-User@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user
>
--
Gökhan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-user/attachments/20091117/2d246242/attachment.html
More information about the SciPy-User
mailing list