Apply suggestions from code review

src/signalcorr.jl

@@ -302,7 +302,7 @@ function crosscov!(r::AbstractMatrix, x::AbstractVector, y::AbstractMatrix, lags
     return r
 end
 
-function crosscov!(r::AbstractArray{U,3} where U, x::AbstractMatrix, y::AbstractMatrix, lags::AbstractVector{<:Integer}; demean::Bool=true)
+function crosscov!(r::AbstractArray{<:Any}, x::AbstractMatrix, y::AbstractMatrix, lags::AbstractVector{<:Integer}; demean::Bool=true)
     lx = size(x, 1)
     nx = size(x, 2)
     ny = size(y, 2)
@@ -458,7 +458,7 @@ function crosscor!(r::AbstractMatrix, x::AbstractVector, y::AbstractMatrix, lags
     return r
 end
 
-function crosscor!(r::AbstractArray{U,3} where U, x::AbstractMatrix, y::AbstractMatrix, lags::AbstractVector{<:Integer}; demean::Bool=true)
+function crosscor!(r::AbstractArray{<:Any}, x::AbstractMatrix, y::AbstractMatrix, lags::AbstractVector{<:Integer}; demean::Bool=true)
     lx = size(x, 1)
     nx = size(x, 2)
     ny = size(y, 2)

src/toeplitzsolvers.jl

@@ -3,25 +3,24 @@
 
 Solve Yule-Walker equations using Durbin algorithm.
 
-Solution of NxN system is obtained iteratively, by solving 1x1, 2x2, ...
+Solution of N×N system is obtained iteratively, by solving 1×1, 2×2, ...
 in succession. For use in computing partial autocorrelation matrix,
-return the last elements of the successive solutions in vector p[].
+return the last elements of the successive solutions in vector `p`.
 
-The section 4.7 of Golub,VanLoan "Matrix Computations," 4th ed.,
-discusses the algorithm in detail.
+Reference: Golub, G. H., and C. F. Van Loan. "Matrix computations 4th edition the johns hopkins university press." Baltimore, MD (2013), section 4.7
 
 # Arguments
-- `r::AbstractVector`: first column of a herimitian positive definite
+- `r::AbstractVector`: first column of a Herimitian positive definite
     Toeplitz matrix, excluding the diagonal element (equal to one).
 - `y::AbstractVector`: work vector for solution, should have the same length
-    as r[]
+    as `r`
 - `p::AbstractVector`: last elements of the successive solutions, should
-     have the same length  as r[]
+     have the same length  as `r`
 
 # Returns
-- `y::AbstractVector`: same as the second argument
+- `y::AbstractVector`: return the solution vector (same as the second argument)
 """
-function durbin!(r::AbstractVector{T}, y::AbstractVector{T}, p::AbstractVector{T}) where T
+function durbin!(r::AbstractVector{T}, y::AbstractVector{T}, p::AbstractVector{T}) where {T}
     n = length(r)
     n <= length(p) || n <= length(y) || throw(DimensionMismatch("Auxiliary vectors cannot be shorter than data vector"))
     y[1] = -r[1]
@@ -48,7 +47,7 @@ function durbin!(r::AbstractVector{T}, y::AbstractVector{T}, p::AbstractVector{T
     end
     return y
 end
-durbin(r::AbstractVector{T}) where T = durbin!(r, zeros(T, length(r)), zeros(T, length(r)))
+durbin(r::AbstractVector{T}) where {T} = durbin!(r, zeros(T, length(r)), zeros(T, length(r)))
 
 function levinson!(r::AbstractVector{T}, b::AbstractVector{T}, x::AbstractVector{T}) where T<:BlasReal
     n = length(b)

test/runtests.jl

@@ -15,6 +15,7 @@ tests = ["ambiguous",
          "hist",
          "rankcorr",
          "signalcorr",
+         "signalcorr_real",
          "misc",
          "pairwise",
          "reliability",

test/signalcorr_complex.jl → test/signalcorr.jl

@@ -29,6 +29,8 @@ cmplx_x2 = convert(AbstractVector{Complex}, x2)
 
 @test autocov([1:5;])          ≈ [2.0, 0.8, -0.2, -0.8, -0.8]
 @test autocor([1, 2, 3, 4, 5]) ≈ [1.0, 0.4, -0.1, -0.4, -0.4]
+@test_throws MethodError autocov([1 missing 2 3 4 5])
+@test_throws MethodError autocor([1 missing 2 3 4 5])
 
 acovx1 =  [0.755284179631112 + 0.0im,
            -0.005333112170365584 - 0.18633291805458002im,
@@ -90,6 +92,7 @@ c22 = crosscov(x2, x2)
 @test crosscov(cmplx_x1, cmplx_x)  ≈ [c11 c12]
 @test crosscov(x,  x)  ≈ cat([c11 c21], [c12 c22], dims=3)
 @test crosscov(cmplx_x,  cmplx_x)  ≈ cat([c11 c21], [c12 c22], dims=3)
+@test_throws MethodError crosscov([1 6; 2 7; missing 8; 3 9; 4 10; 5 11])
 
 rcor0 = [0.230940107675850,
         -0.230940107675850,
@@ -116,7 +119,7 @@ c22 = crosscor(x2, x2)
 @test crosscor(cmplx_x1, cmplx_x)  ≈ [c11 c12]
 @test crosscor(x,  x)  ≈ cat([c11 c21], [c12 c22], dims=3)
 @test crosscor(cmplx_x, cmplx_x)  ≈ cat([c11 c21], [c12 c22], dims=3)
-
+@test_throws MethodError crosscor([1 6; 2 7; missing 8; 3 9; 4 10; 5 11])
 
 ## pacf least squares
 pacf_ls = [-1.598495044296996e-03 - 2.915104118351207e-01im
@@ -134,7 +137,7 @@ function yulewalker_qr(v::AbstractVector)
     x = -A\b
 end
 function toeplitz(v::AbstractVector{T}) where T
-    N=length(v)
+    N = length(v)
     A = zeros(T, N - 1, N - 1)
     for n in 1:N-1
         A[n, n+1:end] = conj(v[2:N-n])