Switching cost models as hypothesis tests

Abstract

 An inference problem with a penalty for mistakes and switching leads to a band of inaction  The band of inaction is the same as the band where you fail to reject H 0 , in a hypothesis test  Hypothesis tests now have a new micro-foundation *Highlights (for review)

Belief adjustment: a double hurdle model and experimental evidence

“…The effect of the questionnaire score is negative and significant, though it is not large. The negative coefficient is consistent with Cohen et al's (2019) model if subjects have cognitive costs. This gives us our first result: Result 1 There is evidence of time-dependent (random) belief adjustment.…”

“…If f ( ) has most probability mass between 0 and 1, most agents only partially adjust, and subjects converge to full adjustment at = 1 to the extent that the probability mass in f ( ) converges towards unity. Cohen et al (2019) show that models with cost-based state-dependent sticky belief adjustment are equivalent to an inferential expectations (IE) model, where agents' hypothesis testing generates infrequent belief adjustment (Menzies and Zizzo 2009). 14 We show this in our specific context in "Appendix 2: Relationship between inferential expectations and switching cost models".…”

“…We believe that a setting where there is no such nudge is a more accurate reflection of many real world decision settings, such as investment portfolio choice over time (Cohen et al 2019). Cost-based state-dependent sticky belief adjustment can be micro-founded on adjustment costs, for example inattention costs (Alvarez et al 2016), information costs (Abel et al 2013), or cognitive costs (Magnani et al 2016); these, in turn, can be modelled in a stylized way using a hypothesis testing framework, as shown by Cohen et al (2019) and explained in "Appendix 2: Relationship between inferential expectations and switching cost models". We avoid incentivizing the distractor task to help with the interpretability of the findings in a first experiment studying the effect of inattention on belief updating; changing this could be an interesting direction of future research.…”

“…Within macroeconomic research, sticky belief adjustment can be seen as a possible microfoundation of sticky price adjustment, for example as a result of inattention and observation costs (Alvarez et al 2016), information costs (Abel et al 2013), cognitive costs (Magnani et al 2016) and the consultation of experts by inattentive agents (Carroll 2003), or some combination of these factors (Cohen et al 2019). State-dependence in beliefs implies a dependence of belief adjustment on the economic state, which in turn may depend on new information flowing in.…”

Cost‐efficient monitoring of continuous‐time stochastic processes based on discrete observations

“…There are several possible strategies to incorporate the learning process into the current modeling framework. The filtering formalism 89,90 can be incorporated into the model as well. Another possibility is to enhance the current model with the Bayesian updating formalism harmonizing with stochastic optimal control problems 91 …”