Auctions: Structural Estimation
We have been covering a variety of different auction types in class, but have yet to nail down which one a seller should choose. The attached research paper sheds some light on this topic using the idea of “Structural Estimation”. This idea states that in order for the seller to produce the maximum profit from an auction, the seller would need to know the individual valuation of the bidders; but since this level of knowledge incredibly hard to achieve, an estimation of the data is needed to fill its place. This strategy works by assuming that for any given auction type, the observed bids come about from the Nash Equilibrium of that auction type. Researchers will use this information estimate the distribution of valuations for the bidders in the future. With this knowledge, the estimation of values can then be compared across all action types to find the one that produces optimal revenue (first price, descending price, second price, or ascending price).
This research paper walks hand in hand with what we have been learning in class. The paper combines knowledge of auctions and Nash Equilibrium to create insightful information that has a real world use: maximizing revenue. Students could even test out the validity of Structural Estimation given a small sample of data from an online resource. Most importantly, the paper shows that everything we are learning in Networks are not strictly separate. Graph Theory, Game Theory, Nash Equilibrium, Auctions and so on can all be joined together to grant a deeper understanding of a network then any one section alone.
Source: http://www.worldwidejournals.in/ojs/index.php/ijsr/article/viewFile/8926/9005