Title :
Applications of stochastic complexity and related computational experiments
Author :
Baikovicius, Jimmy ; Gerencsér, László
Author_Institution :
Dept. of Electr. Eng., McGill Univ., Montreal, Que., Canada
Abstract :
The authors show how to solve model selection and change-point detection problems for ARMA processes in real time, using a form of predictive stochastic complexity. The effect of parameter uncertainty versus model order uncertainty is presented. The proposed solutions are illustrated by means of computer simulations
Keywords :
modelling; statistical analysis; time series; ARMA processes; change-point detection; computational experiments; model order uncertainty; model selection; parameter uncertainty; predictive stochastic complexity; Algebra; Automation; Computer simulation; Difference equations; Particle measurements; Polynomials; Predictive models; Stochastic processes; Uncertain systems; Uncertainty;
Conference_Titel :
Decision and Control, 1992., Proceedings of the 31st IEEE Conference on
Conference_Location :
Tucson, AZ
Print_ISBN :
0-7803-0872-7
DOI :
10.1109/CDC.1992.371025
