Home › Survival › Survival estimators

Runs the two-sample test on every pair of groups: lifelines' pairwise_logrank_test.

public static IReadOnlyList<PairwiseLogRankResult> Pairwise(ReadOnlySpan<double> durations, ReadOnlySpan<int> groups, ReadOnlySpan<bool> eventObserved, LogRankOptions options = null)

Parameters — durations holds one non-negative duration per subject and eventObserved its flag. groups holds one label per subject, any integers, at least two distinct. options chooses the weighting, its exponents and the truncation; null runs the log-rank test.

Returns — one PairwiseLogRankResult per pair of labels, the labels ascending and each pair lower label first: lifelines' row order.

Exceptions — ArgumentException when the spans differ in length, the sample is empty, a duration is negative or NaN, fewer than two groups appear, or a pair's distinct durations times two are more cells than one array holds; each pair is tested on its own, so the groups-by-groups table MultiGroup refuses is never built. ArgumentOutOfRangeException when options names no weighting, or holds a negative or non-finite exponent, or a negative or NaN truncation.

Example — the three groups of the multi-group page, two at a time.

using Lodestar.Survival;

double[] durations = [5, 8, 12, 3, 9, 15, 4, 6, 20, 11, 7, 14];
int[] groups = [1, 1, 1, 2, 2, 2, 3, 3, 3, 1, 2, 3];
bool[] observed = [true, true, false, true, true, true, true, true, false, true, false, true];

IReadOnlyList<PairwiseLogRankResult> pairs = LogRank.Pairwise(durations, groups, observed);

int count = pairs.Count;                                    // => 3
int second = pairs[1].GroupB;                               // => 3
double statistic = Math.Round(pairs[1].Result.Statistic, 6);   // => 0.087424

Remarks — each pair is tested on its own two groups alone: the other groups are out of its risk sets and, under Fleming-Harrington, out of its pooled curve. No correction for multiple comparisons is applied, as lifelines applies none; with many groups, scale the p-values before reading them. Subject weights are not taken, as lifelines' pairwise test takes none.

Applies to — net10.0, netstandard2.0.

See also — LogRank.MultiGroup, PairwiseLogRankResult.