Mehmet Hakan Satman

> It is good to have deepest regression estimator referenced with > > `Van Aelst, Stefan, et al. "The deepest regression method." Journal of Multivariate Analysis 81.1 (2002): 138-166.` >...

R already have this functionality and can be used as a reference: [https://cran.r-project.org/web/packages/DepthProc/DepthProc.pdf](https://cran.r-project.org/web/packages/DepthProc/DepthProc.pdf)

Okay @fmyilmaz, very well, so what is your plan? Is it possible to write pure julia code without any dependency of R or C implementations?

Deepest regression (DR) have multiple methods implemented. The catline is only for single exploratory variable and it is not that necessary. As I remember, there is an exact algorithm for...

@tantei3 can I consult your opinions on a subject: Since the regression depth and the deepest regression estimators are difficult to implement and there is a vast amount of legacy...

the package GLMNet is using fortran code. [Here is the link.](https://github.com/JuliaStats/GLMNet.jl) They put the fortran code in directory /deps and just ccall.

yes, it is too much complicated and translating the code is the most secure method. The most important file is the [RegresDepthDeepest.f](https://github.com/cran/mrfDepth/blob/master/src/RegresDepthDeepest.f) and it depends many other fortran files.

Hi @tantei3 , I found a bucket of code in Julia's GitHub repository, in file [here](https://github.com/JuliaLang/julia/blob/master/test/llvmcall.jl): The code ```julia import Base.llvmcall function f(x, y) llvmcall("""%3 = add i64 %1, %0...

@tantei3 how about shipping the fortran code within the package, trying to compile them in the installation process or before loading first time? If the compiler is absent, when the...

Somebody suggested JuliaPackaging/Binary Builder on Twitter. [Link is here](https://github.com/JuliaPackaging/BinaryBuilder.jl)