The r-out-of-n F-system and load-sharing system are very common in industrial engineering. Statistical inference has been developed here for an equal-load sharing r-out-of-n F-system on Birnbaum-Sauders (BS) lifetime distribution. A simulation study is carried out with different parameter values and different censoring rates in order to examine the performance of the proposed estimation method. Moreover, to find maximum likelihood estimates numerically, three methods of finding initial values for the parameters - pseudo complete sample method, Type-II modified moment estimators of BS distribution method and stochastic approximation method - are developed. These three methods are then compared based on the number of iterations and simulation time. Two real data sets and one simulated data set are used for illustrative purposes. Finally, some concluding comments are made including possible
future directions for investigation. / Thesis / Master of Science (MSc)
| Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/21051 |
| Date | January 2017 |
| Creators | Zhou, Yiliang |
| Contributors | Balakrishnan, Narayanaswamy, Davies, Katherine, Statistics |
| Source Sets | McMaster University |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0019 seconds