Statistical Analysis Of Model Under Geometric Distribution

In the life test, the real failure causes of the system are often blocked due to various factors, such as loss of diagnostic tool, data recording error, time limit, and so on. The data obtained under this situation is called masked data. In recent years, many scholars have done a lot of statistical research and analysis on this type of data,mainly focusing on the maximum likelihood estimate, Bayesian estimate and interval estimate according to the various distributions that the unit obeys in a series system or parallel system when masking occurs. This article is a combination of series system and parallel system. While masking occurs, the maximum likelihood estimation and approximate interval estimation of the parameters of the three-unit parallel-series systems and series-parallel systems whose failure rate is a constant or a linear function that passes through the origin of the system are both derived, respectively.The chapter studies the two-unit parallel system, three-unit series system,three-unit parallel system, three-unit series system and three-cell string parallel system. The system unit follows a geometric distribution,when the mask data occurs,it gives the full sample system lifetime maximum likelihood function and the proof for the unique solution for the maximum likelihood function. Then we do the simulation of the life distribution of the system by Matlab. And we also obtain the maximum likelihood estimation of the parameter for the life distribution of the system.