The paper by HIROTUGU AKAIKE discusses the role of the likelihood function in Bayesian statistical inference, emphasizing its importance as the primary source of objectivity in Bayesian methods. The author argues that the subjective interpretation of the Bayes procedure, which often overlooks the likelihood function, leads to conceptual confusions. AKAIKE suggests that the likelihood function should be central to Bayesian methods, as it represents an objective measure of the goodness of a model. The paper also addresses common misconceptions about the subjective theory of probability, such as the rationality of Savage's axioms and the role of parameters in Bayesian modeling. It introduces a general Bayesian modeling approach for linear problems, demonstrating its practical utility through numerical examples, including seasonal adjustment of time series. The paper concludes by highlighting the importance of the likelihood function in Bayesian procedures and its potential for developing useful statistical models.The paper by HIROTUGU AKAIKE discusses the role of the likelihood function in Bayesian statistical inference, emphasizing its importance as the primary source of objectivity in Bayesian methods. The author argues that the subjective interpretation of the Bayes procedure, which often overlooks the likelihood function, leads to conceptual confusions. AKAIKE suggests that the likelihood function should be central to Bayesian methods, as it represents an objective measure of the goodness of a model. The paper also addresses common misconceptions about the subjective theory of probability, such as the rationality of Savage's axioms and the role of parameters in Bayesian modeling. It introduces a general Bayesian modeling approach for linear problems, demonstrating its practical utility through numerical examples, including seasonal adjustment of time series. The paper concludes by highlighting the importance of the likelihood function in Bayesian procedures and its potential for developing useful statistical models.