lucabrugnolini / vectorautoregressions.jl Goto Github PK

Hi Luca,

It's kind of unfortunate that the package and the module have different names: VAR vs VARs. I think the Julia convention is to keep them the same. We cannot rename the module to VAR as that clashes with the function and type VAR. But what about naming the package and module VectorAutoregressions.jl. The name is a bit more descriptive for non-macroeconomists which might also help when registering the package. I also think it's not too long to type once in a using statement.

Let me know what you think. Best,

Benjamin

Hi Luca,

Sorry to bother you again but I tried to use your package to replicate some Cholesky identified IRFs generated using old Matlab codes of mine and I didn't succeed. I then noticed that in the irf_chol() function you have the following line:
https://github.com/lucabrugnolini/VAR.jl/blob/2f0f8e51e32c92494df1598b1095b5718811cf2b/src/VAR.jl#L548
This is puzzling as full(cholfact(A)) just reconstructs the matrix A but I think what you need is the lower diagonal matrix cholfact(A)[:L]. See the following example

A = [2.0 2.0; 2.0 2.0]
Base.LinAlg.Cholesky{Float64,Array{Float64,2}} with factor:
[1.41421 0.0; 1.41421 2.10734e-8]

cholfact(A)[:L]
2×2 LowerTriangular{Float64,Array{Float64,2}}:
 1.41421   ⋅        
 1.41421  2.10734e-8

full(cholfact(A))
2×2 Array{Float64,2}:
 2.0  2.0
 2.0  2.0

If I replace line 548 by mSigma = cholfact(V.Σ)[:L], I can replicate my results. I'm probably missing something and it would be great if you could help me out.

Best, Benjamin

Suggestion: add a method that checks eigenvalues and returns whether VAR is stable.

From browsing the code quickly there's already a function that generates the companion form, so it'll be simple to implement.

If you want, I can add it myself and make a PR