This will be repository for data and code from Schuckers, Lopez and Macdonald Estimation of Player Aging Curves Using Regression and Imputation Annals of Operations Research 325(1): 681-699 (2023)
Here is the link to the arXiv version of this paper: https://arxiv.org/abs/2110.14017
NewAgingModelsTesting.R is a file of functions for generation of simulated player aging data and for fitting the models in this paper.
Fit_Models_NHL_Data.R fits the models in our paper to data for NHL players found in the file "forwards_zppg_missing_long_2022.csv" and outputs the results in the file "Forwards_model_out_071022.csv".
forwards_zppg_missing_long_2022.csv is a file with data on players from the National Hockey League in a long format tabular format with z-scores
Forwards_model_out_071022.csv is the results of fitting the models described in our paper.
testsim_1_1000_0.8_4.R is an example file that generates a specific set of simuated data, fits models to those data and reports the estimated mean aging curve from those models. This code outputs to files, a file "outsingle<_>.csv" which is the data for all players for a single simulation and "outtest<>.csv" which is the Primary function here is playeragingsim. This function also produces outsingle*.csv
NHL_forwards_models_out_071022.csv is a example file that has an instance of the estimates of the 'aging curve' for each of the model discussed in the paper.
forwards_nhl_aging2020_long.csv is a data file where each row is an observed season for a player.
out_sim_all_073121.csv is a realization of 1000 simulated datasets
outsingle062821_0_600_0.4_4q.csv is a single realization of the simulated data
*