Code to accompany the manuscript entitled "Data-driven clustering of infectious disease incidence into age groups".
Authors: Rami Yaari, Amit Huppert and Itai Dattner.
-
util.R: various utility functions.
-
run_sir.R : generates simulated age group incidence data using an SIR model .
-
msclust.R: implementation of the msclust algorithm.
-
msclust_test.R: testing the peformance of the msclust algorithm using Monte-Carlo simulations of age-group incidence data.
-
msclust_test_with_bagging.R: testing the peformance of the msclust+ algorithm (msclust with bagging) using Monte-Carlo simulations of age-group incidence data.
-
gmm_test.R: testing the peformance of the Gaussian mixture model algorithm using Monte-Carlo simulations of age-group incidence data.
-
sigclust2_test.R: testing the peformance of the sigclust2 algorithm using Monte-Carlo simulations of age-group incidence data.
-
run_alpha_tests.R: running tests to assess the type-I error rate of the various algorithms.
-
run_power_tests.R: running tests to assess the power of the various algorithms.
-
covid19_age_group_clustering.R: running the msclust+ algorithm on real incidence data of Covid-19 from Israel, to obtain age-group clustering per epidemic wave and sector of the population. The data used by this code cannot be made available at the current time for privacy considerations.
-
*.RDS: saved results of running 'run_alpha_tests.R' and 'run_power_tests.R', which are presented in the manuscript tables.
For support and bug reports send an email to: ramiyaari@gmail.com or open an issue [here].