MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Estimation of Mean and Variance of a lognormal distribution from a sample data

Asked by Ashish on 12 Jun 2013

I have a dataset of a certain variable (100 values).

I want to find the mean and the variance of the dataset considering it follows a Lognormal distribution.

At last, I want to generate a random number from the Mean and Variance (found in the previous step).

Products

No products are associated with this question.

Answer by Matt J on 12 Jun 2013

Formulas for the maximum likelihood estimate of the mean and variance are given here:

http://en.wikipedia.org/wiki/Log-normal_distribution#Maximum_likelihood_estimation_of_parameters

Matt J on 12 Jun 2013

Well, for example, the formula for the mean would be

` mu_hat= mean(log(yourData));`
Tom Lane on 12 Jun 2013

This would be an estimate of the mu parameter, but that's not the mean of the lognormal. So if you really want the mean rather than mu, look elsewhere on that wikipedia page and see how to compute the mean from your mu and sigma estimates.

Ashish on 19 Jun 2013

I was able to accomplish from a document for my application. Here is the link-http://engineer.jpl.nasa.gov/practices/at2.pdf

Thanks for the help. At least your points gave me a start.