OPTIMAL SEARCH FOR THE BEST ALTERNATIVE

OPTIMAL SEARCH FOR THE BEST ALTERNATIVE

January 1978 | Martin L. Weitzman
This paper by Martin L. Weitzman characterizes the optimal strategy for searching among multiple sources with different characteristics to find the best outcome. The optimal strategy is based on a reservation price rule, where each source is assigned a reservation price, and the next source to be sampled is the one with the highest reservation price. The stopping rule is to terminate the search when the maximum sampled reward exceeds the reservation price of every unsampled source. This simple rule is derived and interpreted, providing a clear and intuitive economic interpretation of the optimal policy. The paper also discusses the application of this model to various economic search problems, such as job search, price searching, and technology development, and highlights the limitations of the model in more complex scenarios. The proof of optimality is provided, showing that Pandora's rule is indeed the optimal strategy under the assumptions made.This paper by Martin L. Weitzman characterizes the optimal strategy for searching among multiple sources with different characteristics to find the best outcome. The optimal strategy is based on a reservation price rule, where each source is assigned a reservation price, and the next source to be sampled is the one with the highest reservation price. The stopping rule is to terminate the search when the maximum sampled reward exceeds the reservation price of every unsampled source. This simple rule is derived and interpreted, providing a clear and intuitive economic interpretation of the optimal policy. The paper also discusses the application of this model to various economic search problems, such as job search, price searching, and technology development, and highlights the limitations of the model in more complex scenarios. The proof of optimality is provided, showing that Pandora's rule is indeed the optimal strategy under the assumptions made.
Reach us at info@study.space
[slides and audio] Optimal search for the best alternative