Score tests for zero-inflated Poisson models
Score tests for zero-inflated Poisson models
复制标题
DOI：
10.1016/s0167-9473(01)00104-9
复制
发表时间：
2002-07-28
影响因子：
1.8
通讯作者：
Hinde, JP
中科院分区：
文献类型：
--
作者：
Jansakul, N;Hinde, JP
在许多情况下，计数数据有很大比例的零，零膨胀泊松回归（ZIP）模型可能是合适的。van den Brock （biomeics, 51(1995) 738-743）给出了一个简单的零膨胀分数检验，将具有恒定超额零比例的ZIP模型与标准泊松回归模型进行比较。我们将这个检验扩展到更一般的情况，允许零概率依赖于协变量。通过仿真研究对该测试的性能进行了评估。为了确定零通货膨胀模型中潜在的重要协变量，提出了一种复合检验。本文以文献中的两个例子说明了综合分数测试和综合程序的使用。综合分数检验建议了合适的模型。(C) 2002 Elsevier Science B.V.版权所有
In many situations count data have a large proportion of zeros and the zero-inflated Poisson regression (ZIP) model may be appropriate. A simple score test for zero-inflation, comparing the ZIP model with a constant proportion of excess zeros to a standard Poisson regression model, was given by van den Brock (Biometrics, 51 (1995) 738-743). We extend this test to the more general situation where the zero probability is allowed to depend on covariates. The performance of this test is evaluated using a simulation study. To identify potentially important covariates in the zero-inflation model a composite test is proposed. The use of the general score test and the composite procedure is illustrated on two examples from the literature. The composite score test is found to suggest appropriate models. (C) 2002 Elsevier Science B.V. All rights reserved.