Advertising investments on television: real option estimation through Markov chains

We propose the valuation of a real option in the telecommunications industry. According to the probabilistic present worth approach, we estimate the value of a contract between a television network and a company willing to advertise its business on this network. We assume that the value of the contract depends on a time-dependent variable, i.e., the number of viewers tuned into the network, which behaves like a Markov process. After discretizing and converting this number into a monetary value through a specific function, we compute the nth-order moment of the total discounted earnings. The knowledge of the moments, and the application of the maximum-entropy approach, allows to find the probability distribution of the payoff function and the consequential pricing of the real option. Finally, we apply the proposed model to the real television audience data.