论文部分内容阅读
针对某型系列弹密集度评定,给出了基于混合倒伽玛分布的Bayes区间估计方法。该方法通过引入继承因子,有效利用了历史样本信息。通过对不同总体下随机抽样的方式,比较了该方法与经典统计理论中区间估计算法在风险特征方面的差异,结果表明:该方法估计精度更高,具有更好的稳健性,尤其是纳伪概率控制较好,为科学、有效评定系列化弹药密集度提供了思路。
For a series of projectile intensity evaluation, Bayesian interval estimation method based on mixed inverse gamma distribution is given. The method effectively utilizes the historical sample information by introducing inheritance factors. The differences between the method and the classical statistical theory in terms of risk characteristics are compared by means of random sampling under different population. The results show that the proposed method has higher estimation accuracy and better robustness, Probability control is better, providing a scientific and effective assessment of the series of ammunition-intensive ideas.