A Survival Analysis of the Duration of Olympic Records

TL;DR

This paper applies advanced survival analysis techniques to study the duration of Olympic records, comparing models to predict future record-breaking events at the 2012 Olympics.

Contribution

It introduces the use of recurrent-events survival analysis, including logistic models, to analyze Olympic record durations and predict future record-breaking occurrences.

Findings

Logistic model best fits the data according to AIC

Covariate significance identified for the models

Predictions made for records at 2012 Olympics

Abstract

We use recurrent-events survival analysis techniques and methods to analyze the duration of Olympic records. The Kaplan-Meier estimator is used to perform preliminary tests and recurrent event survivor function estimators proposed by Wang & Chang (1999) and Pena et al. (2001) are used to estimate survival curves. Extensions of the Cox Proportional Hazards model are employed as well as a discrete-time logistic model for repeated events to estimate models and quantify parameter significance. The logistic model was the best fit to the data according to the Akaike Information Criterion (AIC). We discuss, in detail, covariate significance for this model and make predictions of how many records will be set at the 2012 Olympic Games in London. Keywords: survival analysis, recurrent events, Kaplan-Meier estimator, Cox proportional hazards model, Olympics.