Class RankingEngine

java.lang.Object

com.quantfinlib.screener.RankingEngine

public final class RankingEngine
extends Object

Ranking engine: scores stocks by a weighted blend of min-max-normalized criteria and sorts them best-first.

Why normalize before blending: raw metrics live on wildly different scales (ROE ~0.15, market cap ~1e11), so a weighted sum of raw values is just "whichever metric has the biggest units wins". Min-max normalization maps each criterion to [0,1] across the candidate set first; the weights then express genuine relative importance. Two consequences to design around. Scores are relative to this run's universe (the same stock scores differently in a different candidate list: fine for "pick the best 20 today", wrong for tracking one name through time). Min-max is outlier-sensitive (one absurd P/E compresses everyone else's spread; screen out garbage with FundamentalFilters before ranking, which is the intended pipeline order). Negative weights invert a criterion (lower P/E ranks higher) without a separate "ascending" flag.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final record	RankingEngine.ScoredStock

Constructor Summary

Constructors

Constructor	Description
RankingEngine()	Creates an instance; the explicit form of the default constructor.

Method Summary

Modifier and Type	Method	Description
RankingEngine	addCriterion(String name, double weight, ToDoubleFunction<StockSnapshot> extractor)	Adds a ranking criterion.
List<RankingEngine.ScoredStock>	rank(List<StockSnapshot> stocks)

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

RankingEngine

public RankingEngine()

Creates an instance; the explicit form of the default constructor.

Method Details

addCriterion

public RankingEngine addCriterion(String name,
 double weight,
 ToDoubleFunction<StockSnapshot> extractor)

Adds a ranking criterion.

Parameters:

weight - relative importance; use a negative weight to prefer smaller values (e.g. lower P/E ranks higher)

extractor - metric to score, e.g. s -> s.fundamentals().roe()

rank

public List<RankingEngine.ScoredStock> rank(List<StockSnapshot> stocks)