Fix mvlognorm

[chrisroadmap]

Pull request

Please confirm that this pull request has done the following:

• [x] Tests added
• [x] Documentation added (where applicable)
• [x] Example added (either to an existing notebook or as a new notebook, where applicable)
• [x] Description in CHANGELOG.rst added

Update to the ensemble generator to use the variance of the input data rather than the standard deviation. Compare the original R code: https://rdrr.io/cran/MethylCapSig/src/R/mvlognormal.R

[codecov[bot]]

Codecov Report

Merging #57 into master will increase coverage by 1.09%. The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #57      +/-   ##
==========================================
+ Coverage   90.78%   91.88%   +1.09%     
==========================================
  Files          33       33              
  Lines        1270     1269       -1     
==========================================
+ Hits         1153     1166      +13     
+ Misses        117      103      -14

Impacted Files	Coverage Δ
fair/tools/ensemble.py	94.33% <100%> (+25.82%)	:arrow_up:

---

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 9e5b197...f3c3d2f. Read the comment docs.

[znicholls]

@chrisroadmap the one other thing that might be worth putting in somewhere, not sure where, is that the ensemble generated here is different from a formal posterior predictive distribution (what most people call a 'probabilistic distribution'), because it's derived by filtering results, rather than through a Monte Carlo Markov Chain or other likelihood calculation.

[chrisroadmap]

That might be getting too technical (I'll admit I'm not up to speed with the terminology here) but feel free to PR if you can explain it :)