PortfolioExperiment

Models & Types · AiDotNet.Finance.Trading.Evaluation

One experiment in the greenfield trading-agent research: a named environment configuration to train + evaluate. Every SOTA-relevant axis is a knob here — objective (Reward), leverage budget (MaxLeverage), and market frictions (Frictions) — so the options become experiments ranked on the holdout rather than an either/or pick. (Policy/algorithm is the agent factory’s axis.)

Implements: IEquatable<PortfolioExperiment>

Constructors

Constructor Summary
PortfolioExperiment(string Name, IPortfolioReward Reward, double MaxLeverage, PortfolioFrictions Frictions) One experiment in the greenfield trading-agent research: a named environment configuration to train + evaluate.

Properties

Property Summary
IPortfolioReward Reward { get; set; }
PortfolioFrictions Frictions { get; set; }
double MaxLeverage { get; set; }
string Name { get; set; }

Methods

Method Summary
PortfolioExperiment <Clone>$()
PortfolioManagerEnvironment<T> BuildEnvironment(IReadOnlyList<double[]> assetPrices, IReadOnlyList<double[]> featureColumns, int windowSize, double initialCapital) Builds the environment this experiment specifies over the given data.
bool Equals(PortfolioExperiment other)
bool Equals(object obj)
bool op_Equality(PortfolioExperiment left, PortfolioExperiment right)
bool op_Inequality(PortfolioExperiment left, PortfolioExperiment right)
int GetHashCode()
string ToString()
void Deconstruct(string Name, IPortfolioReward Reward, double MaxLeverage, PortfolioFrictions Frictions)