Study On The Method Of Determining The Minimum Threshold Of Sample Size In Two-parameter Exponential Distribution Test

Hypothesis test is one of the main contents of statistical inference.In financial insurance,medical treatment,industry and other aspects,the loss caused by making the type error in the testing process is sometimes much greater than the loss caused by making the type error,and the affects the estimation effect of the population directly.Most scholars have studied the control of two types of errors and the determination of sample size in the hypothesis test of normal distribution.However,the research in controlling two types of errors and determining the sample size in the hypothesis test of two-parameter exponential distribution is few,which is one of the representative distributions in financial insurance,medical treatment and industry.Based on this phenomena,the method of determining the sample size which can control the probability of making two types of errors at the same time is explored.The formula of determining the minimum sample size is obtained by constructing appropriate test statistics and exploring the quantitative relationship between the probability of two types of errors and the sample size.The minimum sample size and the efficiency of controlling two types of errors are obtained by numerical simulation.From the results,we can find that the probability of making two types of errors is no more than any given upper limit value after the maximum permissible absolute(relative)error is given.