Key points are not available for this paper at this time.
We present a general optimal stopping problem accommodating a broad spectrum of nonstationary environments. These include scenarios where the decision maker's patience (i. e. , discount rate), time pressure (i. e. , arrival rate of a stochastic deadline), learning speed (i. e. , volatility of the diffusion process) can change over time both gradually and abruptly. Our paper offers three main contributions. First, we prove this general problem has a well-defined solution under mild regularity conditions. Second, we develop comprehensive comparative statics results that are crucial to characterize the shape of the stopping region in the broad class of monotone environments. Finally, we leverage these comparative statics to examine the speed-accuracy tradeoffs in various information acquisition problems, revealing how decision accuracy varies over time in response to changes in the discount rate, learning speed, or deadline-induced time pressure. Notably, our main comparative static results hold locally and thus can also capture non-monotone relations.
Durandard et al. (Sat,) studied this question.