Benedikt Venn

The same applies to UniProt documentation

add nelder mead c3a98d412c2f58d992d5ea0b46995b9092bae86e

I think the best way to model the support is to use `FSharp.Stats.Inverval

Totally agree to the ofTuple and toTuple member:+1: I'm not sure if in any case it is allowed to have endValue < startValue. I dont't think so and therefore I...

Additionally we need confidence and prediction bands for qq plots: see: `stat_qq_band` and `The Power to See: A New Graphical Test of Normality,Aldor-Noiman, 2013`

Setting unique Axis ids seems to fix this behaviour. Details ```csharp #r "nuget: Plotly.NET, 3.0.0" #r "nuget: Plotly.NET.CSharp, 0.8.0" #r "nuget: Plotly.NET.Interactive, 3.0.0" using Plotly.NET.CSharp; var testData = new[]{1.0,1.1,2.0}; var...

I totally agree to remove the AutoOpen tags. However I'm not sure if `[]` is necessary, as I have no in-depth experience using it. Although I don't see any problems...

## Summary As you mentioned there are two result types of a QR decomposition: (i) full/complete form QR (Householder) and (ii) reduced form QR (Gram-Schmidt)[1](https://de.wikipedia.org/wiki/QR-Zerlegung#Definition),[2](https://qph.cf2.quoracdn.net/main-qimg-c19b21f42d093963b26bf5784c8aeafb),[3](https://studyflix.de/mathematik/qr-zerlegung-1786). Gram-Schmidt seems to be efficient...

The normal InvCDF for `mean = 0` and `sigma = 1` is already implemented at an inproper position: https://github.com/fslaborg/FSharp.Stats/blob/b74ecf29651b32f66e328958931cdb2d2dd8dc0f/src/FSharp.Stats/Signal/QQPlot.fs#L91-L92 I agree, we should add quantile functions as `InvCDF` for all...

I've added an InvCDF member to all distributions by 3d6a2201c45d792d95cc127c97c377fa4bcf496c. I noticed the approximation of the inverse error function leads to some discrepancies when extreme values are chosen and compared...