NelsonAalenEstimator<T>

Models & Types · AiDotNet.SurvivalAnalysis

Implements the Nelson-Aalen estimator for non-parametric cumulative hazard function estimation.

Inheritance: SurvivalModelBase<T> → NelsonAalenEstimator<T>

Implements: IGeneratedParameterRegistrar<T>

For Beginners

The Nelson-Aalen estimator calculates the cumulative hazard function H(t), which represents the accumulated risk of an event up to time t. Unlike Kaplan-Meier which estimates survival probability S(t), Nelson-Aalen estimates H(t) directly. They’re related by S(t) = exp(-H(t)).

How It Works

How it works:

  • At each event time t, add d(t)/n(t) to the cumulative hazard
  • d(t) = number of events at time t
  • n(t) = number of subjects at risk just before time t

Variance estimation: The Nelson-Aalen estimator uses the variance formula: Var(H(t)) = sum over event times of d(t)/n(t)^2

When to use:

  • When you want to estimate cumulative hazard directly
  • As input to other models that work with cumulative hazard
  • When comparing hazard rates across groups

Reference: Nelson (1972), Aalen (1978)

Example

// age and treatment arm per patient
var features = new Matrix<double>(new double[,]
{
    { 45, 1 }, { 52, 0 }, { 38, 1 }, { 61, 0 }, { 47, 1 }, { 55, 0 }
});
// months each patient was observed
var times = new Vector<double>(new double[] { 5.0, 12.0, 3.0, 18.0, 9.0, 21.0 });
// 1 = the event happened; 0 = censored, still fine when the study ended or lost to follow-up
var events = new Vector<int>(new int[] { 1, 0, 1, 0, 1, 1 });
            
var result = new AiModelBuilder<double, Matrix<double>, Vector<double>>()
    .ConfigureModel(new NelsonAalenEstimator<double>())
    .Build(features, times, events);
            
var risk = result.Predict(features);

Constructors

Constructor Summary
NelsonAalenEstimator() Creates a new Nelson-Aalen estimator.

Properties

Property Summary
Vector<T> CumulativeHazard { get; } Gets the cumulative hazard function values at event times.
Vector<T> Variance { get; } Gets the variance estimates at event times.

Methods

Method Summary
IFullModel<T, Matrix<T>, Vector<T>> WithParameters(Vector<T> parameters)
Matrix<T> PredictCumulativeHazard(Vector<T> times, Matrix<T> features) Predicts cumulative hazard at specified times (override to use direct estimate).
Matrix<T> PredictSurvivalProbability(Matrix<T> x, Vector<T> times) Predicts survival probabilities at specified times.
Vector<T> GetBaselineSurvival(Vector<T> times) Gets the baseline survival function.
Vector<T> Predict(Matrix<T> input) Predicts survival probability at median event time.
Vector<T> PredictHazardRatio(Matrix<T> x) Returns hazard ratios (all 1s for non-parametric estimator).