Authors: Anna Bykhovskaya (Assistant Professor at Duke University)
Abstract: In this talk we will discuss the dynamic Tobit model, which can be used to model censored time series. We will start with characterizing its asymptotic behavior. Turns out, the asymptotics has a point of discontinuity (analogue of a unit root for linear autoregression). We will then extend local to unity asymptotics to the non-linear setting of the dynamic Tobit model, motivated by this discontinuity as well as by the application of this model to highly persistent censored time series. We will present a unit root test in the presence of censoring, and we will provide an application of our methods to the Swiss franc / euro exchange rate, during a period when this was subject to an occasionally binding lower bound.