Estimation of the number of failures in the Weibull model using the ordinary differential equation
Estimation of the number of failures in the Weibull model using the ordinary differential equation
复制标题
使用常微分方程估计威布尔模型中的故障数
DOI：
10.1016/j.ejor.2012.07.011
复制
发表时间：
2012
影响因子：
6.4
通讯作者：
H. Hirose
中科院分区：
文献类型：
--
作者：
Masuda T;Beppu T;Muzumoto T;IshikoT;Chikamoto A;Hayashi H;Okabe H,Otao R;Hasita H;Takamori H;Baba H;H. Hirose
在使用右截断分组数据估计故障数时，我们经常会遇到使用似然原理对条件概率估计值小于真实值的情况。在传染病传播预测中，通常使用联立常微分方程描述的SIR模型，即使在观测数据量较小的情况下，它也可以相当好地预测感染患者的数量。我们研究了在右截断分组数据的情况下，常微分方程模型是否能够比似然原理更准确地估计故障数。与 SARS、A(H1N1) 和 FMD 病例类似，在 Weibull 模型中获得了阳性结果。
In estimating the number of failures using right truncated grouped data, we often encounter cases that the estimate is smaller than the true one when we use the likelihood principle to conditional probability. In infectious disease spread predictions, the SIR model described by simultaneous ordinary differential equations is commonly used, and it can predict reasonably well the number of infected patients even when the size of observed data is small. We have investigated whether the ordinary differential equation model can estimate the number of failures more accurately than does the likelihood principle under the condition of right truncated grouped data. The positive results are obtained in the Weibull model, similarly to the cases of the SARS, A(H1N1), and FMD.