Discontinuous decision processes and threshold autoregressive time series modelling
在贝叶斯决策理论框架下，将不连续决策过程的分析应用于非线性时间序列模型近似，自然推导出门限自回归模型。
Setting the problem of approximating an underlying nonlinear time series model within the framework of Bayesian decision theory, we demonstrate how the general analysis of discontinuous decision processes developed by Smith, Harrison & Zeeman (1981) leads naturally to a threshold autoregression.