Home › Survival › Survival estimators

Fits an accelerated failure time regression to interval-censored times: lifelines' fit_interval_censoring.

public static AftSummary FitIntervalCensored(AftModel model, ReadOnlySpan<double> design, ReadOnlySpan<double> lower, ReadOnlySpan<double> upper, int featureCount, AftOptions options = null)

public static AftSummary FitIntervalCensored(AftModel model, ReadOnlySpan<double> design, ReadOnlySpan<double> lower, ReadOnlySpan<double> upper, ReadOnlySpan<double> weights, ReadOnlySpan<double> entries, int featureCount, AftOptions options = null)

The second overload takes lifelines' weights_col and entry_col.

Parameters — model, design, featureCount and options are Fit's. lower and upper bound each subject's event: lower non-negative, upper at least lower, infinite for a subject still event-free at lower; equal bounds are an event observed exactly. weights holds one positive, finite weight per subject, or is empty for ones; entries each subject's entry time, non-negative and at most its upper bound, or is empty for none.

Returns — an AftSummary, as Fit returns, with a ConcordanceIndex of NaN.

Exceptions — ArgumentException when the spans do not match the subjects, a covariate is not finite, a bound is NaN or negative, an upper bound is below its lower one, a weight or entry span is neither empty nor one valid value per subject, or the primary parameter has no column at all. ArgumentOutOfRangeException when model names no model, or featureCount is negative. InvalidOperationException when the fit does not converge, or its parameters are not identified. ArgumentException also when the parameters are too many for their square to fit in one array (#1311). ArgumentException naming featureCount also when the parameters are more than one array holds, and naming design when the design is more cells than one array holds or its length only matches the subjects times featureCount once that product wraps int (#1614).

Example — visits bracketing each event, against a dose.

using Lodestar.Survival;

double[] dose = [1.0, 2.0, 0.5, 3.0, 0.0, 1.5, 0.0, 2.5, 1.0, 0.5];
double[] lastClear = [2, 4, 0, 6, 3, 8, 5, 1, 7, 10];
double[] firstSeen = [4, 6, 3, 9, 3, double.PositiveInfinity, 7, 4, 10, double.PositiveInfinity];

AftSummary fit = AcceleratedFailureTime.FitIntervalCensored(AftModel.Weibull, dose, lastClear, firstSeen, featureCount: 1);

double perUnit = Math.Round(fit.Coefficients[0], 6);   // => -0.017618
bool unscored = double.IsNaN(fit.ConcordanceIndex);    // => True

Remarks — there is no concordance index for an interval-censored fit. lifelines computes none, and reading its concordance_index_ raises AttributeError; NaN says the same without a throw. A subject contributes S(lower) − S(upper) at its own linear predictors, as in ParametricSurvival.FitIntervalCensored.

Applies to — net10.0, netstandard2.0.

See also — AcceleratedFailureTime, AcceleratedFailureTime.Fit, AftSummary.