Added Maximum Likelihood Estimation for Beta Distribution (fit_mle)

[jg-854]

Included unit tests to check functionality works - ran tests with no errors from my modifications.
Updated documentation too.

[codecov-io]

Codecov Report

Merging #1267 (cc47342) into master (863844c) will increase coverage by 0.04%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1267      +/-   ##
==========================================
+ Coverage   81.29%   81.34%   +0.04%     
==========================================
  Files         117      117              
  Lines        6566     6583      +17     
==========================================
+ Hits         5338     5355      +17     
  Misses       1228     1228              

Impacted Files	Coverage Δ
src/univariate/continuous/beta.jl	70.43% <100.00%> (+5.12%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 863844c...cc47342. Read the comment docs.

[andreasnoack]

Thanks for the PR. However, this has been discussed a couple of times in the past and we are not sure if it's generally a good idea that individual distributions implement their own Newton iterations. Is the Hessian guaranteed to be negative definite?

[jg-854]

Is the Hessian guaranteed to be negative definite?

Yes it is negative definite. I understand your point, a general Newton's method might a be a good idea.