View Item 
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        JavaScript is disabled for your browser. Some features of this site may not work without it.

        Browse

        All of UU Student Theses RepositoryBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

        How to Find the One - Secretary Problem Extended with Uncertain Observations

        Thumbnail
        View/Open
        Thesis_Florijn.pdf (2.610Mb)
        Publication date
        2019
        Author
        Florijn, C.P.
        Metadata
        Show full item record
        Summary
        Making a good decision is often a challenge. How to choose your ideal house? How to find your true love? These problems resemble the mathematical puzzle called the secretary problem. In the classic secretary problem, administrators want to hire the best secretary from n applicants. The secretaries present themselves one by one. Each decision, rejecting or accepting, needs to be immediate, is permanent, but can only be based on the ranking of seen applicants. Who should they hire? The optimal strategy is a cut-off strategy, where the number of applicants one needs to wait is n/e, with n is the number of applicants, which is approximately 37 with n = 100. After 37%, the administrators should hire the first who is better than all they have seen before. In this study, the effect of uncertain observations on the chance of success of cut-off strategies has been evaluated by the use of computer simulations. Computer simulations were used to simulate versions of the secretary problem with intervals, normal distributions and pareto distributions. The trend showed that the larger the uncertainty, the lower the cutoffpoint from the optimal strategy. The reserved advice is: if you are not sure, choose sooner!
        URI
        https://studenttheses.uu.nl/handle/20.500.12932/33168
        Collections
        • Theses
        Utrecht university logo