Sequential Decision Making Predictions under the Influence of Observational Learning

Today’s corporate managers face challenges in information technology (IT) adoption with great stakes. The waiting period for an IT investment to be realized, if any, could be long; thus word-of-mouth information propagation may not help them to make wise decisions. Though information used by early adopters to make their decisions may not be available to the public, late adopters can often observe the decisions made by the early adopters and infer hidden information to supplement their own private information. Observational learning theory applies when a person uses observed behavior from others to infer something about the usefulness of the observed behavior. Walden and Browne proposed a simulation procedure to model the influence of observational learning in sequential decision makings. Previously, we proposed a dynamic Bayesian network (DBN) to model sequential decision makings under the influence of observational learning. In the present study, we show how to infer a DBN model from simulated data. Hidden Markov model and artificial neural networks were used to infer the DBN model. Their performance will be discussed.