CAJ | 학술논문

针对双参数指数分布的截尾数据，利用一种新的参数Bootstrap方法，构建了可靠寿命的置信下限，并从理论上证明了给出的置信下限是精确的。在小样本情形下，通过比较新的参数 Bootstrap方法取得的置信下限与 Engelhardt-Bain方法取得的近似置信下限的覆盖率，发现参数 Bootstrap 置信下限更令人满意。
침대쌍삼수지수분포적절미수거，이용일충신적삼수Bootstrap방법，구건료가고수명적치신하한，병종이론상증명료급출적치신하한시정학적。재소양본정형하，통과비교신적삼수 Bootstrap방법취득적치신하한여 Engelhardt-Bain방법취득적근사치신하한적복개솔，발현삼수 Bootstrap 치신하한경령인만의。
For constructing lower confidence limit of reliability life of two-parameter exponential distribution,we propose a new parametric bootstrap method,which is conceptually simpler and straightforward than other methods previously considered in the literature.It is shown that the interval is exact,that is ,its coverage probability is equal to the required confidence level.For small sample,simulation study shows that compared to the approximate interval due to En-gelhardt-Bain,the coverage probability of the confidence interval based on the parametric boot-strap method is more satisfactory.