# Tune Curve to Histogram

4 answers

If you know the underlying distribution (i.e. skewed Gaussian, etc.), you can manually perform maximum likelihood estimates for the distribution parameters and then plot the resulting distribution on top of the histogram. However, you need to normalize your histogram so that you can see empirical probabilities instead of numbers.

0

source to share

I think you want it to fit a distribution, not a curve, which may not have a finite area under the curve. The data looks like it is being censored on the right tail, but it may well follow a normal log distribution or gamma distribution throughout. If you have a statistics toolbar, try `gamfit`

or `lognfit`

for a starter.

0

source to share