An essential ingredient of any time series anatysis is the estimation of the modcl parameters. The main objective of this paper is to develop a convenient Rayesian technique for estimation which can be used to analyze ‘seasonal autoregressive moving average processes. The foundation of the proposed approach is to approximate the conditional likelihood by a normal-gamma distribution on the parameter space; Based on the approximated conditional likelihood function, the marginal posterior distribution of the coefficients of the model is approximated by a t distribu- tion, and the marginal posterior distribution of the model precision is approximated by a gamma distribution. The proposed technique is illustrated by some numerical examples.
Shaarawy, S., & Ismail, M. (1987). Bayesian Inference for Seasonal ARMA Models. The Egyptian Statistical Journal, 31(1), 77-103. doi: 10.21608/esju.1987.428903
MLA
Samir Shaarawy; Mohamed Ali Ismail. "Bayesian Inference for Seasonal ARMA Models", The Egyptian Statistical Journal, 31, 1, 1987, 77-103. doi: 10.21608/esju.1987.428903
HARVARD
Shaarawy, S., Ismail, M. (1987). 'Bayesian Inference for Seasonal ARMA Models', The Egyptian Statistical Journal, 31(1), pp. 77-103. doi: 10.21608/esju.1987.428903
VANCOUVER
Shaarawy, S., Ismail, M. Bayesian Inference for Seasonal ARMA Models. The Egyptian Statistical Journal, 1987; 31(1): 77-103. doi: 10.21608/esju.1987.428903
)
View on SCiNiTO