Replaced GammaMethod.cov_AD with ADerrors.cov_sym

src/fit_functions.jl

@@ -88,7 +88,7 @@ function fit_routine(model::Function,
     [uwerr(y, wpm) for y in ydata]
     if length(W) ==0
         if corr
-            C = length(C) ==0 ? GammaMethod.cov_AD(ydata) : C
+            C = length(C) ==0 ? ADerrors.cov_sym(ydata) : C
             W = LinearAlgebra.pinv(C);
             W = (W'+W)*0.5
         else
@@ -102,11 +102,14 @@ function fit_routine(model::Function,
     chi2 = sum(fit.resid.^2)
     par = fit_error(chisq,LsqFit.coef(fit),ydata,wpm,W,false)
     # if the fit is correlated  we have already C, no need to recompute it.
-    chiexp,pval =  if length(C) !=0
-        GammaMethod.chiexp(chisq,LsqFit.coef(fit),value.(ydata),C,W),pvalue(chisq,chi2,LsqFit.coef(fit),ydata,W,C=C)
-    else
-        ADerrors.chiexp(chisq,LsqFit.coef(fit),ydata,wpm,W=W),pvalue(chisq,chi2,LsqFit.coef(fit),ydata,W,wpm=wpm)
-    end
+    chiexp  = ADerrors.chiexp(chisq,
+                              LsqFit.coef(fit),
+                              ydata,
+                              wpm,
+                              C=length(C)==0 ? nothing : C,
+                              W=W)
+
+    pval = pvalue(chisq,chi2,LsqFit.coef(fit),ydata,W,wpm=wpm,C=C)
 
     compute_error_for_logging(logfile,par);
     println(logfile, "fit_routine output:")

src/pvalue.jl

@@ -35,7 +35,7 @@ end
     W::Union{Vector{Float64},Array{Float64,2}} = Vector{Float64}();
     wpm = Dict{Int64,Vector{Float64}}()
     nmc::Int64 = 5000,
-    C::AbstractMatrix{Float64} = GammaMethod.cov_AD(data,wpm))
+    C::AbstractMatrix{Float64} = Matrix{Float64}(undef,0,0))
 
 Computes the p-value of a previously done fit, using as input the `\chi^2` observed from the fit, the fit parameters and the fitted data.
 The p-value for a given `\chi^2` is the probability of, given the data you have, finding such a `\chi^2` or worse from a fit, and still
@@ -67,7 +67,7 @@ function pvalue(chisq::Function,
     W::AbstractVector{Float64}=Vector{Float64}();
     wpm::Union{Dict{Int64,Vector{Float64}},Dict{String,Vector{Float64}}, Nothing} =  Dict{Int64,Vector{Float64}}(),
     nmc::Int64 = 5000,
-    C::AbstractMatrix{Float64} = ADerrors.cov(data,wpm))
+    C::AbstractMatrix{Float64} = Matrix{Float64}(undef,0,0))
 
     n = length(xp)   # Number of fit parameters
     m = length(data) # Number of data
@@ -128,6 +128,8 @@ function pvalue(chisq::Function,
         _Px
     end
 
+    length(C)!=0 || (C=ADerrors.cov_sym(data))
+
     nu = let
         aux = sqrt(C)
         aux2 = similar(aux)