Regularisation Constraints

PortfolioOptimisers.AbstractRegularisationEstimator Type

abstract type AbstractRegularisationEstimator <: AbstractEstimator

Abstract supertype for all portfolio weight regularisation estimators.

Related

• LpRegularisation

source

PortfolioOptimisers.LpRegularisation Type

struct LpRegularisation{__T_p, __T_val} <: AbstractRegularisationEstimator

Lp-norm regularisation term added to the optimisation objective:

$$\text{penalty}=\mathrm{val}\cdot {(\sum _{i=1}^N|w_i{|}^p)}^{1/p}$$

Penalises concentrated portfolios by encouraging weight smoothness for $p>1$.

Fields

• p::Number: Lp norm exponent. Must satisfy p > 1 and be finite.

• val::Number: Penalty coefficient. Must be positive and finite.

Constructors

LpRegularisation(; p::Number = 3, val::Number = 1e-3) -> LpRegularisation

Related

• AbstractRegularisationEstimator

• AbstractRegularisationEstimator

source