Removed Laplace from the list of fit.md

[Dsantra92]

As requested in #807, I have removed the Laplace from the list of distribution from the list in fit.md .

[Dsantra92]

On another note Laplace supports fit_mle(dist, x) and does not support fit_mle(dist, x, w). I would like to ask if I could add the fit_mle(dist,x,w) for Laplace in a future PR.

[andreasnoack]

I don't think removing it from the list is the way to go here. As you mention, fit_mle(dist, x) works so the right thing seems to be to clarify which of the distributions support the weight option in fit_mle

[Dsantra92]

@andreasnoack Can I just make a PR to add weighted_fit_mle for Laplacian ? I guess that will resolve the issue.

[andreasnoack]

Yes. That would be even better.

[Dsantra92]

Laplace weighted fit_mle requires weighted MAD and I found no implementation in StatsBase.jl. I implemented my own MAD and realized it's too messy for a single function to holdv(which is not even intended to it). I actually see 3 options:

1. Implement weighted MAD in StatsBase.jl and then use it.
2. Implement weighted map in the fit_mle function itself.
3. Or as you suggested, just mention that the weighted fit_mle is not implemented in the package.

I believe 3 would be a quick fix for now and I will add the fit_mle with MAD in a follow-up pr.

[codecov-commenter]

Codecov Report

Merging #1187 (c4e4f3a) into master (e127e9d) will decrease coverage by 0.38%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1187      +/-   ##
==========================================
- Coverage   81.89%   81.51%   -0.39%     
==========================================
  Files         116      115       -1     
  Lines        6534     6626      +92     
==========================================
+ Hits         5351     5401      +50     
- Misses       1183     1225      +42     

Impacted Files	Coverage Δ
src/matrixvariates.jl	76.81% <0.00%> (-14.10%)	⬇️
src/multivariates.jl	70.66% <0.00%> (-10.92%)	⬇️
src/mixtures/mixturemodel.jl	68.11% <0.00%> (-10.49%)	⬇️
src/univariates.jl	62.20% <0.00%> (-5.35%)	⬇️
src/univariate/continuous/logitnormal.jl	64.28% <0.00%> (-4.95%)	⬇️
src/univariate/discrete/hypergeometric.jl	63.41% <0.00%> (-3.26%)	⬇️
src/univariate/discrete/skellam.jl	81.25% <0.00%> (-2.63%)	⬇️
src/univariate/continuous/pgeneralizedgaussian.jl	61.76% <0.00%> (-1.88%)	⬇️
src/univariate/discrete/soliton.jl	90.90% <0.00%> (-1.69%)	⬇️
src/common.jl	67.56% <0.00%> (-1.67%)	⬇️
... and 41 more

---

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 e127e9d...c4e4f3a. Read the comment docs.

[devmotion]

IIRC this was fixed recently.

[devmotion]

Ah no, the unweighted fit was fixed (#1309) and a PR for the weighted fit was closed (#1310) but is planned to be updated (possibly in a new PR) after JuliaStats/StatsBase.jl#687 is fixed. So depending on how timely the weighted fit is implemented, the warning in this PR might not be needed.