Covariances and correlations

Author: nalimilan

docs/Project.toml

@@ -1,5 +1,6 @@
 [deps]
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
+Statistics = "20745b16-79ce-11e8-11f9-7d13ad32a3b2"
 
 [compat]
 Documenter = "0.27"

docs/src/cov.md

@@ -1,11 +1,23 @@
 # Covariances and Correlations
 
-Functions to computing various types of correlations and covariances are provided.
+Functions to computing various types of covariances and correlations are provided.
 
 ## Covariance, Correlation and Scatter Matrix
+
 ```@docs
 cov
 cor
+scattermat
+cov2cor
+cor2cov
+CovarianceEstimator
+SimpleCovariance
+```
+
+## Partial Correlation
+
+```@docs
+partialcor
 ```
 
 ## Autocovariance and Autocorrelation

src/Statistics.jl

@@ -13,7 +13,7 @@ using Base: has_offset_axes, require_one_based_indexing
 
 using Printf: @printf
 
-export cor, cov, std, stdm, var, varm, mean!, mean,
+export std, stdm, var, varm, mean!, mean,
     median!, median, middle, quantile!, quantile,
     # moments.jl
     skewness, kurtosis,
@@ -22,12 +22,24 @@ export cor, cov, std, stdm, var, varm, mean!, mean,
     weights, aweights, fweights, pweights,
     # scalarstats.jl
     geomean, harmmean, genmean, mode, modes, percentile, span, variation, sem, mad, mad!,
-    iqr, genvar, totalvar, entropy, renyientropy, crossentropy, kldivergence, describe
-
+    iqr, genvar, totalvar, entropy, renyientropy, crossentropy, kldivergence, describe,
+    zscore, zscore!,
+    # cov.jl
+    cor, cov, scattermat, cov2cor, cor2cov, CovarianceEstimator, SimpleCovariance,
+    # partialcor.jl
+    partialcor,
+    # signalcorr.jl
+    autocov!, autocov, autocor!, autocor, crosscov!, crosscov, crosscor!, crosscor,
+    pacf!, pacf
+
+include("common.jl")
 include("weights.jl")
 include("wsum.jl")
 include("moments.jl")
 include("scalarstats.jl")
+include("cov.jl")
+include("partialcor.jl")
+include("signalcorr.jl")
 
 ##### mean #####
 

src/common.jl

@@ -25,17 +25,4 @@ const RealFP = Union{Float32, Float64}
 fptype(::Type{T}) where {T<:Union{Float32,Bool,Int8,UInt8,Int16,UInt16}} = Float32
 fptype(::Type{T}) where {T<:Union{Float64,Int32,UInt32,Int64,UInt64,Int128,UInt128}} = Float64
 fptype(::Type{Complex{Float32}}) = Complex{Float32}
-fptype(::Type{Complex{Float64}}) = Complex{Float64}
-
-# A convenient typealias for deprecating default corrected Bool
-const DepBool = Union{Bool, Nothing}
-
-function depcheck(fname::Symbol, b::DepBool)
-    if b == nothing
-        msg = "$fname will default to corrected=true in the future. Use corrected=false for previous behaviour."
-        Base.depwarn(msg, fname)
-        false
-    else
-        b
-    end
-end
+fptype(::Type{Complex{Float64}}) = Complex{Float64}

src/cov.jl

@@ -63,18 +63,6 @@ matrix (`corrected=false`) is computed by multiplying `scattermat(X, w)` by
 """
 cov
 
-
-"""
-    mean_and_cov(x, [wv::AbstractWeights,] vardim=1; corrected=false) -> (mean, cov)
-
-Return the mean and covariance matrix as a tuple. A weighting
-vector `wv` can be specified. `vardim` that designates whether
-the variables are columns in the matrix (`1`) or rows (`2`).
-Finally, bias correction is applied to the covariance calculation if
-`corrected=true`. See [`cov`](@ref) documentation for more details.
-"""
-function mean_and_cov end
-
 scattermat(x::DenseMatrix; mean=nothing, dims::Int=1) =
     _scattermatm(x, mean, dims)
 _scattermatm(x::DenseMatrix, ::Nothing, dims::Int) =
@@ -91,11 +79,11 @@ _scattermatm(x::DenseMatrix, wv::AbstractWeights, mean, dims::Int) =
 
 ## weighted cov
 covm(x::DenseMatrix, mean, w::AbstractWeights, dims::Int=1;
-     corrected::DepBool=nothing) =
+     corrected::Bool=true) =
     rmul!(scattermat(x, w, mean=mean, dims=dims), varcorrection(w, depcheck(:covm, corrected)))
 
 
-cov(x::DenseMatrix, w::AbstractWeights, dims::Int=1; corrected::DepBool=nothing) =
+cov(x::DenseMatrix, w::AbstractWeights, dims::Int=1; corrected::Bool=true) =
     covm(x, mean(x, w, dims=dims), w, dims; corrected=depcheck(:cov, corrected))
 
 function corm(x::DenseMatrix, mean, w::AbstractWeights, vardim::Int=1)
@@ -113,16 +101,6 @@ Compute the Pearson correlation matrix of `X` along the dimension
 cor(x::DenseMatrix, w::AbstractWeights, dims::Int=1) =
     corm(x, mean(x, w, dims=dims), w, dims)
 
-function mean_and_cov(x::DenseMatrix, dims::Int=1; corrected::Bool=true)
-    m = mean(x, dims=dims)
-    return m, covm(x, m, dims, corrected=corrected)
-end
-function mean_and_cov(x::DenseMatrix, wv::AbstractWeights, dims::Int=1;
-                      corrected::DepBool=nothing)
-    m = mean(x, wv, dims=dims)
-    return m, cov(x, wv, dims; corrected=depcheck(:mean_and_cov, corrected))
-end
-
 """
     cov2cor(C, s)
 

test/runtests.jl

@@ -891,3 +891,6 @@ end
 include("weights.jl")
 include("wsum.jl")
 include("moments.jl")
+include("cov.jl")
+include("partialcor.jl")
+include("signalcorr.jl")