Hi,
It seems that some methods are missing to compute the firsts 2 (weighted) moments of complex random variables.
Is any conceptual issues that prevents me to propose a PR implementing these?
julia> N = 1_000_000;
julia> x = randn(N);
julia> xc = randn(ComplexF64,N);
julia> w = rand(N);
julia> mean(x)
-0.0006234813354142627
julia> mean(xc)
0.0002818700630158139 + 0.0004029044695052171im
julia> var(x)
0.9994220449190467
julia> var(xc)
1.000575477440655
julia> moment(xc,1)
ERROR: MethodError: no method matching moment(::Vector{ComplexF64}, ::Int64)
The function `moment` exists, but no method is defined for this combination of argument types.
Closest candidates are:
moment(::AbstractArray{<:Real}, ::Int64)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:314
moment(::AbstractArray{<:Real}, ::Int64, ::Real)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:314
moment(::AbstractArray{<:Real}, ::Int64, ::AbstractWeights)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:321
...
Stacktrace:
[1] top-level scope
@ REPL[59]:1
julia> moment(x,1)
-1.2931877790833823e-17
julia> wsum(xc,w)
180.43932062446382 + 205.28102880933926im
julia> mean(xc,weights(w))
0.00036099221296754424 + 0.00041069126515038817im
julia> var(xc,weights(w))
ERROR: MethodError: no method matching var(::Vector{ComplexF64}, ::Weights{Float64, Float64, Vector{Float64}})
The function `var` exists, but no method is defined for this combination of argument types.
Closest candidates are:
var(::CovarianceEstimator, ::AbstractVector; kwargs...)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/cov.jl:296
var(::AbstractArray{<:Real}, ::AbstractWeights, ::Int64; mean, corrected)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:63
var(::AbstractArray; corrected, mean, dims)
@ Statistics ~/.julia/packages/Statistics/gbcbG/src/Statistics.jl:380
...
Stacktrace:
[1] top-level scope
@ REPL[63]:1
julia> mean_and_var(x, weights(w))
(0.00035486592111588897, 0.9988627381401368)
julia> mean_and_var(xc, weights(w))
ERROR: MethodError: no method matching mean_and_var(::Vector{ComplexF64}, ::Weights{Float64, Float64, Vector{Float64}})
The function `mean_and_var` exists, but no method is defined for this combination of argument types.
Closest candidates are:
mean_and_var(::Any; corrected)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:117
mean_and_var(::AbstractArray{<:Real}, ::AbstractWeights; corrected)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:139
mean_and_var(::AbstractArray{<:Real}, ::AbstractWeights, ::Int64; corrected)
@ StatsBase ~/.julia/packages/StatsBase/lRQEN/src/moments.jl:163
...
Stacktrace:
[1] top-level scope
@ REPL[65]:1
julia> mean_and_var(xc)
(0.0002818700630158139 + 0.0004029044695052171im, 1.000575477440655)
Hi,
It seems that some methods are missing to compute the firsts 2 (weighted) moments of complex random variables.
Is any conceptual issues that prevents me to propose a PR implementing these?