在這篇論文中，我們將 Wen et al. 在2006年提出的貝氏迴歸模型做推廣，考慮其誤差項會受到迴歸函數的函數值所影響。除此之外，我們也更近一步地探討，在每一個位置附近發生跳點的機率以及其所跳的高度之間的關係。 One of the purposes of this thesis is to extend the Bayesian step-function regression model of Wen et al. (2006) to allow for more general noise term so that the error term may vary with the ratio itself, which is also a familiar phenomenon in regression analysis. The other purpose of this thesis is to explore in more detail the relation between the size of the jump at a point of the posterior mode and posterior probability that it is a jump point.
