[SciPy-User] Fitting a curve on a log-normal distributed data

Gökhan Sever gokhansever@gmail....
Tue Nov 17 16:30:56 CST 2009

On Tue, Nov 17, 2009 at 4:03 PM, Lorenzo Isella <lorenzo.isella@gmail.com>wrote:

>  Date: Mon, 16 Nov 2009 23:44:17 -0600
>> From: G?khan Sever <gokhansever@gmail.com>
>> Subject: [SciPy-User] Fitting a curve on a log-normal distributed data
>> To: Discussion of Numerical Python <numpy-discussion@scipy.org>,
>>  SciPy
>>        Users List <scipy-user@scipy.org>
>> Message-ID:
>>        <49d6b3500911162144x1193e04cj1a103776092c4471@mail.gmail.com>
>> Content-Type: text/plain; charset="utf-8"
>> Hello,
>> I have a data which represents aerosol size distribution in between 0.1 to
>> 3.0 micrometer ranges. I would like extrapolate the lower size down to 10
>> nm. The data in this context is log-normally distributed. Therefore I am
>> looking a way to fit a log-normal curve onto my data. Could you please
>> give
>> me some pointers to solve this problem?
>> Thank you.
> Hello,
> I have not followed the many replies to this long post in detail, but by
> chance I happen to know quite in detail what you are talking about (probably
> SMPS data or similar).
> I normally resort to R for this kind of tasks (http://www.r-project.org/),
> but nothing prevents you from using Python instead. You just want to compare
> your empirical data binning with what would be expected from a lognormal
> distribution. Please have a look at
> http://tinyurl.com/ygmw4lc
> and at the functions defined there (A1, mu1 and myvar1 are the overall
> concentration, the geometric mean and the std of the number-size
> distribution, respectively).
> Cheers

Hey Lorenzo,

Finally someone who knows the heart of the subject :) Thanks for stopping

The data that I am using is Passive Cavity Aerosol Spectrometer (PCASP)
measured size-distributions. Unfortunately even if we had the mains part of
the SMPS instrument we couldn't fly it since the radioactive element was not
reached during campaign. It is always an issue to deliver the radioactive
parts out of the US :)

Anyways assuming that the relative humidity was quite low in the measurement
region I am not expecting a huge deviation from the dry-particle size
definition. But as I said above this is just a feasibility study. I will
test and see how much an error I will get with this method. Besides there is
no information regarding to the chemical composition of the aerosols,
therefore I am basing on kappa-kohler theory and making another
simplification at that point.

Could you please send this script off-list and the data file associated with

Would be greatly appreciated.

> Lorenzo

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-user/attachments/20091117/b457f37b/attachment-0001.html 

More information about the SciPy-User mailing list