H
Hozumi Morohosi
Researcher at National Graduate Institute for Policy Studies
Publications - 15
Citations - 92
Hozumi Morohosi is an academic researcher from National Graduate Institute for Policy Studies. The author has contributed to research in topics: Markov chain Monte Carlo & Hybrid Monte Carlo. The author has an hindex of 5, co-authored 15 publications receiving 86 citations. Previous affiliations of Hozumi Morohosi include University of Tokyo.
Papers
More filters
Book ChapterDOI
A Practical Approach to the Error Estimation of Quasi-Monte Carlo Integrations
Hozumi Morohosi,Masanori Fushimi +1 more
TL;DR: A simple error estimation method is used for quasi-Monte Carlo integrations, and some theoretical considerations on the errors given by these two methods are given.
Journal ArticleDOI
Applying the shortest‐path‐counting problem to evaluate the importance of city road segments and the connectedness of the network‐structured system
Tatsuo Oyama,Hozumi Morohosi +1 more
TL;DR: A quantitative method for evaluating the stable connectedness of the network-structured system using shortest-path-counting methods and the application of the Monte Carlo method to estimate the expected stable-connection function is proposed.
Journal ArticleDOI
Measuring the network robustness by Monte Carlo estimation of shortest path length distribution
TL;DR: Monte Carlo methods for the computation of two robustness measure for networks on the connectivity of randomly chosen pair of vertices and the shortest path length between pair of connected vertices are devised.
A case study of optimal ambulance location problems
TL;DR: A comparison of those optimization models through actual patient call data from Tokyo metropolitan area is given to show the characteristics of each model and investigate a possibility of improvement in ambulance service.
Journal ArticleDOI
Applying path‐counting methods for measuring the robustness of the network‐structured system
TL;DR: This work considers the path-counting problem that asks how many paths exist between any two different nodes in a network after deleting an arbitrary number of edges or nodes from the original network and proposes a quantitative method for measuring the robustness of the network-structured system.