In this paper, statistical inference for a competing risks model is discussed when latent failure times belong to a general family of inverted exponentiated distributions. Based on a generalized progressive hybrid censored data with partially observed failure causes, estimations for unknown parameters are presented under nonrestricted and restricted parameter cases from classic and Bayesian perspectives, respectively. The existence and uniqueness of maximum likelihood estimators of the unknown parameters are established, and the associated approximate confidence intervals are also constructed via Fisher information matrix. In sequel, the Bayes estimators and credible intervals of the parameters are also obtained as well. Finally, the performance of different estimators are evaluated using Monte Carlo simulations and a real data set is also analyzed for illustration.

在本文中，当潜在故障时间属于逆指数分布的一般族时，讨论了竞争风险模型的统计推断。基于具有部分观察到的故障原因的广义渐进混合删失数据，分别从经典和贝叶斯角度在非受限和受限参数情况下呈现对未知参数的估计。建立了未知参数最大似然估计量的存在唯一性，并通过Fisher信息矩阵构造了相关的近似置信区间。随后，也得到了参数的贝叶斯估计量和可信区间。最后，