Accept-reject decision in online advertising using Geometric Brownian Motion
Narayanan Kutty, Varun
MetadataShow full item record
Advertising is quickly shifting from being broadcast over TV and other methods to online websites and search engines. There is still a gamut of research work that requires attention to optimize and maximize revenue generated from advertisement. Also advertisement display techniques and optimal use of advertisement space has been a challenge. The decision to accept or reject incoming advertisements needs to be further evaluated. In the research for this particular thesis we develop an accept reject criteria based on an existing PPV advertisement and decide to accept or reject incoming PPV or PPC advertisement. This is formulated using Stochastic Linear Programming techniques. We randomly sample from a normal distribution the click through rate of the advertisements to generate revenue owing to PPC advertisements. Incoming PPV advertisement request is assumed to be the same as the one in hand. We make the assumption that the page-views for the particular website follow Geometric Brownian Motion and perform statistical analysis to test the efficacy of our assumption. We derive results for daily, weekly and monthly assumptions of page-views via statistical analysis. We conclude the thesis by providing opportunities for future research work.