5D Library › Relationship Rupture Survival Analysis, No-Contact Memory Decay and Reconciliation Game Windows · 中文

Chapter 1. Reconciliation as a Survival Problem

Two events compete

After a rupture, two things can happen to a former couple: they can get back together, or one or both can move on, for example into a new relationship. Once someone has moved on, reconciliation is no longer the same question. Studies of on-again, off-again dating find that getting back together is common, not rare (Dailey et al. 2009); the question here is how to count it. Survival analysis is the natural tool, but it has to be used with care when two events compete.

p01 follows a synthetic cohort of 2,000 former couples for a year. Three in ten are ambivalent and reunite quickly if nothing else happens; the rest are settled. Every couple faces the same chance of moving on.

Calculation

# Chapter 1. p01: Who gets back together, and how to count it. A
# synthetic cohort of 2,000 former couples is followed for 365 days.
# Two events compete: reconciliation, and "moving on" (a new partner,
# after which reconciliation is treated as no longer possible). 30% of
# couples are ambivalent (reconciliation hazard 1/60 a day); the rest
# are settled (1/1500 a day). Moving on has hazard 1/200 a day for all.
# Kaplan-Meier treats moving on as censoring and so overstates the
# chance of reconciliation; the cumulative incidence does not. Seed 92.
import random

N, DAYS, P_AMB, H_AMB, H_SET, H_MOVE = 2000, 365, 0.30, 1 / 60, \
    1 / 1500, 1 / 200

def cohort(rng):
    out = []
    for _ in range(N):
        h = H_AMB if rng.random() < P_AMB else H_SET
        tr, tm = rng.expovariate(h), rng.expovariate(H_MOVE)
        if min(tr, tm) > DAYS:
            out.append((DAYS, "censored"))
        elif tr < tm:
            out.append((tr, "reconciled"))
        else:
            out.append((tm, "moved on"))
    return out

def km_and_ci(data, t_end):
    data = sorted(data)
    at_risk, s_km, s_all, ci = len(data), 1.0, 1.0, 0.0
    for t, kind in data:
        if t > t_end:
            break
        if kind == "reconciled":
            ci += s_all / at_risk      # Aalen-Johansen increment
            s_km *= 1 - 1 / at_risk    # KM: moving on as censoring
        if kind != "censored":
            s_all *= 1 - 1 / at_risk   # survival from either event
        at_risk -= 1
    return 1 - s_km, ci

if __name__ == "__main__":
    rng = random.Random(92)
    data = cohort(rng)
    print("   by day   Kaplan-Meier   cumulative incidence   observed share")
    for t in (30, 90, 180, 365):
        km, ci = km_and_ci(data, t)
        obs = sum(1 for d, k in data if k == "reconciled" and d <= t) / N
        print("   {:6d}   {:12.3f}   {:20.3f}   {:14.3f}".format(
            t, km, ci, obs))
    rec = [d for d, k in data if k == "reconciled"]
    print("   reconciliations in the first 90 days: {:.3f} of all".format(
        sum(1 for d in rec if d <= 90) / len(rec)))
    print("   couples who moved on within the year: {:.3f}".format(
        sum(1 for d, k in data if k == "moved on") / N))
#    by day   Kaplan-Meier   cumulative incidence   observed share
#        30          0.137                  0.128            0.128
#        90          0.286                  0.241            0.240
#       180          0.379                  0.290            0.289
#       365          0.456                  0.311            0.310
#    reconciliations in the first 90 days: 0.775 of all
#    couples who moved on within the year: 0.597

The usual Kaplan-Meier estimate (Kaplan and Meier 1958) treats a couple that moved on as “censored”, as if they might still reunite later. It reports 0.137 by day 30, 0.286 by day 90 and 0.456 by the end of the year. But the share that really reunited is only 0.128, 0.240 and 0.310. The cumulative incidence (Aalen and Johansen 1978), which counts moving on as an event that removes the couple from the question, gives 0.128, 0.241, 0.290 and 0.311, in line with what happened. The gap grows with time because 0.597 of couples moved on within the year.

The same cohort also shows why “most reunions happen early” is true and misleading at once: 0.775 of all reunions came in the first 90 days. That is a statement about who reunites (mostly the ambivalent couples, who do so quickly), not a promise that waiting is useless for anyone in particular.

Does the window close?

A second popular claim is that the chance of reunion “closes” after a few months. p02 builds a world where it cannot: each couple keeps a constant daily chance of reuniting for the whole year. The couples simply differ in how large that chance is.

Calculation

# Chapter 1. p02: Does the "window" close, or do the likely couples
# simply leave first? Every couple keeps a constant daily hazard of
# reconciling for the whole year (no couple's chance ever declines),
# but couples differ: hazards are drawn from a gamma law with mean
# 1/250 a day and shape k. The hazard seen in the surviving group still
# falls month by month, because the couples most likely to reconcile
# leave the group first. For k = 0.5 the decline is steep. Seed 92.
import random

def life_table(rng, k, n=20000, days=360):
    hs = [rng.gammavariate(k, (1 / 250) / k) for _ in range(n)]
    ts = [rng.expovariate(h) for h in hs]
    rows = []
    for a in range(0, days, 60):
        risk = [t for t in ts if t >= a]
        ev = sum(1 for t in risk if t < a + 60)
        rows.append((a, ev / len(risk) / 60 * 1000))
    return rows

if __name__ == "__main__":
    rng = random.Random(92)
    print("   observed hazard per 1,000 couples per day, by window")
    print("   window (days)   all alike   shape 2.0   shape 0.5")
    tabs = [life_table(rng, k) for k in (1e6, 2.0, 0.5)]
    for i in range(6):
        a = tabs[0][i][0]
        print("   {:3d} to {:3d}      {:9.2f}   {:9.2f}   {:9.2f}".format(
            a, a + 59, tabs[0][i][1], tabs[1][i][1], tabs[2][i][1]))
    f = tabs[2][-1][1] / tabs[2][0][1]
    print("   shape 0.5: last window / first window = {:.3f}".format(f))
#    observed hazard per 1,000 couples per day, by window
#    window (days)   all alike   shape 2.0   shape 0.5
#      0 to  59           3.62        3.38        2.96
#     60 to 119           3.56        3.11        2.14
#    120 to 179           3.57        2.83        1.77
#    180 to 239           3.55        2.65        1.40
#    240 to 299           3.55        2.48        1.19
#    300 to 359           3.58        2.17        1.05
#    shape 0.5: last window / first window = 0.355

When all couples are alike the observed hazard stays between 3.55 and 3.62 per 1,000 couples per day in every window. When they differ, the hazard seen in the group still falls month by month: with a strongly mixed population (shape 0.5) it starts at 2.96 and ends at 1.05, so the last window is only 0.355 of the first. Nobody’s own chance changed. The likeliest couples reunited first and left the group, so the couples still apart are, on average, the less likely ones. A falling curve on a chart is therefore not evidence that time itself closes a window; it may only show who is left.

Hypothesis · not established

The author’s interpretation (not yet tested): most of the decline in reconciliation rates over the months after a rupture reflects differences between couples rather than a decline in each couple’s own chance. What would show it wrong: in a cohort followed with repeated measures of each couple’s mutual contact and stated wish to reunite, the within-couple chance of reunion falls with time even after those measures are held fixed, by about as much as the group hazard falls.

Summary

Reconciliation should be counted with the competing event of moving on, or the chance of reunion is overstated; in p01 the overstatement by the end of the year is almost half as large again as the real share. A hazard that falls with time can come entirely from selection. Both points matter for anyone who reads statistics about “how many couples get back together” and applies them to themselves.

References

Aalen, Odd O., and Søren Johansen. 1978. “An Empirical Transition Matrix for Non-Homogeneous Markov Chains Based on Censored Observations.” Scandinavian Journal of Statistics 5 (3): 141 to 150.
Dailey, René M., Abigail Pfiester, Borae Jin, Gary Beck, and Glenn Clark. 2009. “On-Again/off-Again Dating Relationships: How Are They Different from Other Dating Relationships?” Personal Relationships 16 (1): 23 to 47.
Kaplan, E. L., and Paul Meier. 1958. “Nonparametric Estimation from Incomplete Observations.” Journal of the American Statistical Association 53 (282): 457 to 481.

Fan, Y. (2026). Relationship Rupture Survival Analysis, No-Contact Memory Decay and Reconciliation Game Windows: Negative-Affect Dissipation Rates, Dual-Process Nostalgia Peaks and Signaling Equilibria [关系断裂生存分析、断联期记忆衰减与复联博弈窗口] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090241 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).