Asset selection

Asset selectors narrow the universe from the data: drop constant columns, keep the best or worst assets by a risk measure, prune redundant ones. They are ordinary returns-preprocessing estimators — they know nothing about pipelines, and a Pipeline drives them through fit_preprocessing and apply_preprocessing like any other step.

The universe a selector chooses on the training window is its fitted state. Applying the fitted result to an unseen window replays that universe rather than re-deciding it, which is what makes a selector safe inside cross-validation.

See docs/adr/0029-asset-selection-is-returns-preprocessing.md for the design rationale, and PortfolioOptimisers.AbstractAssetSelector for the seam every selector shares.

Scoring assets with a risk measure

A ScoreSelector scores each asset by evaluating a risk measure on that asset's own return series, then hands the scores to a rule. Any risk measure whose supports_precomputed_returns is true may be used, which covers the quantile and drawdown families, the moment measures, and MeanReturn. bigger_is_better tells the ordinal rules which end of the ordering is "best".

Two measures are notable exceptions. Variance and StandardDeviation are WeightsInput measures: their functors consume portfolio weights, not a return series, so they cannot score a single asset and are rejected at construction. Use SCM(), which computes the same quantity from a return series — ZeroVarianceFilter spells this for you.

PortfolioOptimisers.ScoreSelector Type

julia

struct ScoreSelector{__T_score, __T_rule} <: AbstractAssetSelector

Asset selector that scores every asset with a risk measure and keeps the assets a rule admits.

score is any AbstractBaseRiskMeasure that can be evaluated on a bare return series — asset i's score is score(X[:, i]). That reuses the whole risk-measure family: ConditionalValueatRisk and the drawdown measures score risk, SCM() scores variance, MeanReturn scores mean return. bigger_is_better tells the ordinal rules which end is "best".

rule decides what to do with the scores: an absolute band (ThresholdRule) or a count/fraction taken from the tails (RankRule, QuantileRule).

The selected universe is fitted state, so a ScoreSelector is safe inside cross-validation: assets are chosen on the training window and the same universe is replayed on test windows.

Fields

score: Risk measure scoring each asset's return series (AbstractBaseRiskMeasure).
rule: Rule mapping the scores to a keep-mask (AbstractSelectionRule).

Constructors

julia

ScoreSelector(;
    score::AbstractBaseRiskMeasure,
    rule::AbstractSelectionRule,
) -> ScoreSelector

Keywords correspond to the struct's fields.

Validation

supports_precomputed_returns(score). Variance and StandardDeviation are WeightsInput measures and are rejected with a pointer to SCM().

Examples

julia

julia> rd = ReturnsResult(; nx = ["A", "B", "C"], X = [0.1 0.0 -0.2; -0.1 0.0 0.3; 0.2 0.0 -0.1]);

julia> sel = ScoreSelector(; score = SCM(), rule = ThresholdRule(; lo = 1e-12));

julia> PortfolioOptimisers.fit_preprocessing(sel, rd).nx
2-element Vector{String}:
 "A"
 "C"

Related

Asset selection ​

Scoring assets with a risk measure ​

Selection rules ​

Discarding redundant assets ​

Internals ​

Asset selection

Scoring assets with a risk measure

Selection rules

Discarding redundant assets

Internals