TY - GEN
T1 - Evolutionary algorithms applied to likelihood function maximization during poisson, logistic, and Cox proportional hazards regression analysis
AU - Peterson, Leif E.
N1 - Publisher Copyright:
© 2014 IEEE.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2014/9/16
Y1 - 2014/9/16
N2 - Metaheuristics based on genetic algorithms (GA), covariance matrix self-adaptation evolution strategies (CMSA-ES), particle swarm optimization (PSO), and ant colony optimization (ACO) were used for minimizing deviance for Poisson regression and maximizing the log-likelihood function for logistic regression and Cox proportional hazards regression. We observed that, in terms of regression coefficients, CMSA-ES and PSO metaheuristics were able to obtain solutions that were in better agreement with Newton-Raphson (NR) when compared with GA and ACO. The rate of convergence to the NR solution was also faster for CMSA-ES and PSO when compared with ACO and GA. Overall, CMSA-ES was the best-performing method used. Key factors which strongly influence performance are multicollinearity, shape of the log-likelihood gradient, and positive definiteness of the Hessian matrix.
AB - Metaheuristics based on genetic algorithms (GA), covariance matrix self-adaptation evolution strategies (CMSA-ES), particle swarm optimization (PSO), and ant colony optimization (ACO) were used for minimizing deviance for Poisson regression and maximizing the log-likelihood function for logistic regression and Cox proportional hazards regression. We observed that, in terms of regression coefficients, CMSA-ES and PSO metaheuristics were able to obtain solutions that were in better agreement with Newton-Raphson (NR) when compared with GA and ACO. The rate of convergence to the NR solution was also faster for CMSA-ES and PSO when compared with ACO and GA. Overall, CMSA-ES was the best-performing method used. Key factors which strongly influence performance are multicollinearity, shape of the log-likelihood gradient, and positive definiteness of the Hessian matrix.
UR - http://www.scopus.com/inward/record.url?scp=84908569507&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84908569507&partnerID=8YFLogxK
U2 - 10.1109/CEC.2014.6900660
DO - 10.1109/CEC.2014.6900660
M3 - Conference contribution
AN - SCOPUS:84908569507
T3 - Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014
SP - 1054
EP - 1061
BT - Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2014 IEEE Congress on Evolutionary Computation, CEC 2014
Y2 - 6 July 2014 through 11 July 2014
ER -