PortfolioOptimisers.jl

Appearance

6

Processed covariance ​

PortfolioOptimisers.ProcessedCovariance Type
julia
ProcessedCovariance

A covariance estimator that applies a custom matrix processing algorithm and positive definite projection to the output of another covariance estimator. This type enables flexible post-processing of covariance matrices by first computing a base covariance, then applying a user-specified matrix processing algorithm and positive definiteness correction in sequence.

Fields

  • ce: The underlying covariance estimator to be processed.

  • alg: The matrix processing algorithm to apply to the covariance matrix.

  • pdm: The positive definite matrix projection method.

Constructors

julia
ProcessedCovariance(; ce::AbstractCovarianceEstimator = Covariance(),
                    alg::Option{<:AbstractMatrixProcessingAlgorithm} = nothing,
                    pdm::Option{<:Posdef} = Posdef())

Keyword arguments correspond to the fields above.

Examples

julia
julia> ProcessedCovariance()
ProcessedCovariance
   ce ┼ Covariance
      │    me ┼ SimpleExpectedReturns
      │       │   w ┴ nothing
      │    ce ┼ GeneralCovariance
      │       │   ce ┼ StatsBase.SimpleCovariance: StatsBase.SimpleCovariance(true)   
      │       │    w ┴ nothing
      │   alg ┴ Full()
  alg ┼ nothing
  pdm ┼ Posdef
      │      alg ┼ UnionAll: NearestCorrelationMatrix.Newton
      │   kwargs ┴ @NamedTuple{}: NamedTuple()

Related

source
Statistics.cov Method
julia
cov(ce::ProcessedCovariance, X::MatNum; dims = 1, kwargs...)

Compute the processed and positive definite projected covariance matrix for the data matrix X using the specified ProcessedCovariance estimator.

Arguments

  • ce: The ProcessedCovariance estimator specifying the base covariance estimator, matrix processing algorithm, and positive definite projection.

  • X: The data matrix (observations × assets).

  • dims: The dimension along which to compute the covariance.

  • kwargs...: Additional keyword arguments passed to the underlying covariance estimator and processing algorithm.

Returns

  • sigma::MatNum: processed covariance matrix.

Validation

  • dims in (1, 2).

Details

  • Computes the covariance matrix using the base estimator in ce.

  • Transposes X if dims == 2 to ensure variables are in columns.

  • Ensures the covariance matrix is mutable.

  • Applies positive definite projection using the method in ce.pdm.

  • Applies the matrix processing algorithm in ce.alg to the covariance matrix.

  • Returns the processed covariance matrix.

Related

source
Statistics.cor Method
julia
cor(ce::ProcessedCovariance, X::MatNum; dims = 1, kwargs...)

Compute the processed and positive definite projected correlation matrix for the data matrix X using the specified ProcessedCovariance estimator.

Arguments

  • ce: The ProcessedCovariance estimator specifying the base covariance estimator, matrix processing algorithm, and positive definite projection.

  • X: The data matrix (observations × assets).

  • dims: The dimension along which to compute the correlation.

  • kwargs...: Additional keyword arguments passed to the underlying correlation estimator and processing algorithm.

Returns

  • rho::MatNum: processed correlation matrix.

Validation

  • dims in (1, 2).

Details

  • Computes the correlation matrix using the base estimator in ce.

  • Transposes X if dims == 2 to ensure variables are in columns.

  • Ensures the correlation matrix is mutable.

  • Applies positive definite projection using the method in ce.pdm.

  • Applies the matrix processing algorithm in ce.alg to the correlation matrix.

  • Returns the processed correlation matrix.

Related

source