PortfolioOptimisers.jl

Appearance

18

Cardinality Constraints ​

PortfolioOptimisers.set_card_constraints! Function
julia
set_card_constraints!(
    model::Model,
    sp::AbstractMIPSpace,
    ind::AbstractMIPIndicators,
    card::Integer
)

Add the cardinality constraint sum(ib) ≤ card to the JuMP optimisation model.

ind is the indicator bundle the MIP builder that ran returned; the held binary (held_bin) is summed, so the sum counts held assets and the bound caps how many may be non-zero. The constraint key is named for the space sp via mip_key, so the same emitter serves the asset space and every sub-group. The fall-through method does nothing when card is nothing — and so never reads a held indicator the bundle may not carry.

Arguments

  • model::JuMP.Model: The JuMP optimisation model.

  • sp::AbstractMIPSpace: Weight space the constraint acts on.

  • ind::AbstractMIPIndicators: Indicator bundle returned by the MIP builder that ran.

  • card::Option{<:Integer}: Maximum number of non-zero assets, or nothing.

Returns

  • nothing.

Related

source
PortfolioOptimisers.set_gcard_constraints! Function
julia
set_gcard_constraints!(
    model::Model,
    sp::AbstractMIPSpace,
    ind::AbstractMIPIndicators,
    gcard::LinearConstraint
)

Add the group-cardinality constraints A * ib ≤ B and A * ib = B to the model.

Enforces the linear group-cardinality constraint carried by gcard on the held binary (held_bin) of the bundle ind, capping (or fixing) how many assets in each group may be held. The constraint keys are named for the space sp via mip_key, so the same emitter serves the asset space and every sub-group. The fall-through method does nothing when gcard is nothing — and so never reads a held indicator the bundle may not carry.

Arguments

  • model::JuMP.Model: The JuMP optimisation model.

  • sp::AbstractMIPSpace: Weight space the constraints act on.

  • ind::AbstractMIPIndicators: Indicator bundle returned by the MIP builder that ran.

  • gcard::Option{<:LinearConstraint}: Group-cardinality constraint, or nothing.

Returns

  • nothing.

Related

source
PortfolioOptimisers.set_all_smip_constraints! Function
julia
set_all_smip_constraints!(
    model::Model,
    wb::WeightBounds,
    card::Union{Nothing, Integer},
    gcard::Union{Nothing, LinearConstraint},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}},
    lt::Union{Nothing, Threshold},
    st::Union{Nothing, Threshold},
    ss::Union{Nothing, Number}
)
set_all_smip_constraints!(
    model::Model,
    wb::WeightBounds,
    card::Union{Nothing, Integer},
    gcard::Union{Nothing, LinearConstraint},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}},
    lt::Union{Nothing, Threshold},
    st::Union{Nothing, Threshold},
    ss::Union{Nothing, Number},
    i::Integer
)

Add all sub-group MIP constraints for a single or multiple selection matrices.

The single-matrix method runs the shared indicator builder (run_mip_builder!) on a SubsetMIPSpace — applying long/short thresholds and weight bounds — then emits the sub-group cardinality and group-cardinality constraints through the same space-generic emitters as the asset space (set_card_constraints!, set_gcard_constraints!). The vector method iterates over collections of cardinalities, group constraints, and selection matrices.

Arguments

  • model::JuMP.Model: The JuMP optimisation model.

  • wb::WeightBounds: Weight bound specification containing lower and upper bounds.

  • card: Cardinality bound(s) for the sub-group(s).

  • gcard: Group-cardinality constraint(s) for the sub-group(s).

  • smtx: Selection matrix (or vector thereof) for the sub-group(s).

  • lt: Long-side minimum-holding threshold(s).

  • st: Short-side minimum-holding threshold(s).

  • ss::Option{<:Number}: Big-M scaling constant (computed via get_mip_ss when nothing).

  • i::Integer = 1: Index for generating unique names (single-matrix method only).

Returns

  • nothing.

Related

source
PortfolioOptimisers.set_scardmip_constraints! Function
julia
set_scardmip_constraints!(
    model::Model,
    wb::WeightBounds,
    card::Union{Nothing, Integer},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}},
    lt::Union{Nothing, Threshold},
    st::Union{Nothing, Threshold},
    ss::Union{Nothing, Number}
)
set_scardmip_constraints!(
    model::Model,
    wb::WeightBounds,
    card::Union{Nothing, Integer},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}},
    lt::Union{Nothing, Threshold},
    st::Union{Nothing, Threshold},
    ss::Union{Nothing, Number},
    i::Integer
)

Add sub-group cardinality MIP constraints using a selection matrix.

The single-matrix method runs the shared indicator builder (run_mip_builder!) on the sub-group and enforces sum(sib) ≤ card through set_card_constraints!. The vector method iterates over collections of cardinalities and selection matrices.

Arguments

  • model::JuMP.Model: The JuMP optimisation model.

  • wb::WeightBounds: Weight bound specification containing lower and upper bounds.

  • card: Cardinality bound(s) for the sub-group(s).

  • smtx: Selection matrix (or vector thereof).

  • lt: Long-side minimum-holding threshold(s).

  • st: Short-side minimum-holding threshold(s).

  • ss::Option{<:Number}: Big-M scaling constant (computed via get_mip_ss when nothing).

  • i::Integer = 1: Index for generating unique names (single-matrix method only).

Returns

  • nothing.

Related

source
PortfolioOptimisers.set_sgcardmip_constraints! Function
julia
set_sgcardmip_constraints!(
    model::Model,
    wb::WeightBounds,
    gcard::Union{Nothing, LinearConstraint},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}},
    lt::Union{Nothing, Threshold},
    st::Union{Nothing, Threshold},
    ss::Union{Nothing, Number}
)
set_sgcardmip_constraints!(
    model::Model,
    wb::WeightBounds,
    gcard::Union{Nothing, LinearConstraint},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}},
    lt::Union{Nothing, Threshold},
    st::Union{Nothing, Threshold},
    ss::Union{Nothing, Number},
    i::Integer
)

Add sub-group group-cardinality MIP constraints using a selection matrix.

The single-matrix method runs the shared indicator builder (run_mip_builder!) on the sub-group and enforces the linear group cardinality constraints A * sib ≤ B and A * sib = B through set_gcard_constraints!. The vector method iterates over multiple group constraints and selection matrices.

Arguments

  • model::JuMP.Model: The JuMP optimisation model.

  • wb::WeightBounds: Weight bound specification containing lower and upper bounds.

  • gcard: Group-cardinality constraint(s).

  • smtx: Selection matrix (or vector thereof).

  • lt: Long-side minimum-holding threshold(s).

  • st: Short-side minimum-holding threshold(s).

  • ss::Option{<:Number}: Big-M scaling constant (computed via get_mip_ss when nothing).

  • i::Integer = 1: Index for generating unique names (single-matrix method only).

Returns

  • nothing.

Related

source
PortfolioOptimisers.set_smip_constraints! Function
julia
set_smip_constraints!(
    model::Model,
    wb::WeightBounds,
    card::Union{Nothing, Integer, AbstractVector{<:Integer}},
    gcard::Union{Nothing, LinearConstraint, AbstractVector{<:LinearConstraint}},
    smtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}, AbstractVector{<:AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}}},
    sgmtx::Union{Nothing, AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}, AbstractVector{<:AbstractMatrix{<:Union{var"#s34", var"#s33"} where {var"#s34"<:Number, var"#s33"<:AbstractJuMPScalar}}}},
    lt::Union{Nothing, Threshold, AbstractVector{<:Union{Nothing, var"#s823"} where var"#s823"<:Threshold}},
    st::Union{Nothing, Threshold, AbstractVector{<:Union{Nothing, var"#s823"} where var"#s823"<:Threshold}},
    glt::Union{Nothing, Threshold, AbstractVector{<:Union{Nothing, var"#s823"} where var"#s823"<:Threshold}},
    gst::Union{Nothing, Threshold, AbstractVector{<:Union{Nothing, var"#s823"} where var"#s823"<:Threshold}},
    ss::Union{Nothing, Number}
)

Add all sub-group MIP constraints (both cardinality and group-cardinality) to the JuMP optimisation model.

Dispatches between combined selection matrices (calling set_all_smip_constraints!) and separate cardinality/group-cardinality selection matrices (calling set_scardmip_constraints! and set_sgcardmip_constraints! independently).

Arguments

  • model::JuMP.Model: The JuMP optimisation model.

  • wb::WeightBounds: Weight bound specification containing lower and upper bounds.

  • card: Cardinality bound(s).

  • gcard: Group-cardinality constraint(s).

  • smtx: Cardinality selection matrix (or vector thereof).

  • sgmtx: Group-cardinality selection matrix (or vector thereof).

  • lt: Long-side minimum-holding threshold(s) for cardinality sub-groups.

  • st: Short-side minimum-holding threshold(s) for cardinality sub-groups.

  • glt: Long-side minimum-holding threshold(s) for group-cardinality sub-groups.

  • gst: Short-side minimum-holding threshold(s) for group-cardinality sub-groups.

  • ss::Option{<:Number}: Big-M scaling constant (computed via get_mip_ss when nothing).

Returns

  • nothing.

Related

source