Separating Predicted Randomness from Noise

with Miguel Ballester (Oxford) 

Abstract: Given stochastic behavior and a model of stochastic choice, we offer a methodology to separate from the data the randomness that is inherent to the stochastic choice model from what is noisy behavior. We then study the case of several choice models, and apply our methodology to an experimental dataset.

Sign up

More on Jose Apesteguia

Add to my calendar