Linear regression splines are useful tools to describe departures from linearity in several real applications. Location of knots can be seen as change points in the relationship between the variables. In a Bayesian context, we ana- lyze the variation of the Stochastic Search Variable Selection approach previously proposed in Di Credico et al. (2018), focusing on the impact of the hyperparam- eters choice on the estimation of the correct number of knots.
