Quantum Probability and Decisions

Decisions and the description of decision behavior rely heavily on probability theory. Still, there are irrationalities, paradoxes and fallacies when applying models like Rational Choice Theory or Expected Utility Theory. These models are based on classical probability (founded on Kolmogorov’s axioms). Quantum probability is an alternative way to assign probabilities to events. Recent works (see overview in Ashtiani 2015) showed progress in cognition, decision-making and judgment and reasoning by testing if quantum probability models could explain paradoxes and fallacies better than using classical probability. To our knowledge, only goodness-of-fit was evaluated – out-of-sample predictive quality was not tested. We wanted to find out if it is possible to do predictions with quantum probability models. We found that quantum probability models can be used to calculate prediction values in order to evaluate the in-sample and out-of-sample quality of quantum probability models given a specific data set. Some of the results indicate that quantum models are better than classical models and that further exploration is needed. In order to do that, we’ll develop a new model based on quantum probability and evaluate it using the Choice Prediction Competition 2015.