MSP 2016-Calibration of an Optimal Bidding Model for the Mobile Advertisement Markets

One goal of every business is to save money by minimizing cost and maximizing profit. Cidewalk is a mobile advertisement company that wishes to accomplish this goal through implementing an optimal bidding strategy. By collecting and analyzing market data, it was possible to identify the distribution that best represents the second best bids, which is the price Cidewalk pays for an advertisement space by bidding in a Vickrey auction. Implementing a simulation for the optimal bidding model that utilizes Bayesian updating methods ensures that the model responds to any changes in the distribution parameters. The model’s performance was evaluated by simulation and throughout one day the implemented model yielded 17% savings when compared to Cidewalk’s current bidding model.