首页> 外文期刊>Journal of Health Services Research & Policy >Multiple regression of cost data: use of generalised linear models
【24h】

Multiple regression of cost data: use of generalised linear models

机译:成本数据的多元回归:使用广义线性模型

获取原文
获取原文并翻译 | 示例
           

摘要

Objective: Choosing an appropriate method for regression analyses of cost data is problematic because it mustnfocus on population means while taking into account the typically skewed distribution of the data. In this paper wenillustrate the use of generalised linear models for regression analysis of cost data.nMethods: We consider generalised linear models with either an identity link function (providing additive covariateneffects) or log link function (providing multiplicative effects), and with gaussian (normal), overdispersed poisson,ngamma, or inverse gaussian distributions. These are applied to estimate the treatment effects in two randomisedntrials adjusted for baseline covariates. Criteria for choosing an appropriate model are presented.nResults: In both examples considered, the gaussian model fits poorly and other distributions are to be preferred.nWhen there are variables of prognostic importance in the model, using different distributions can materially affectnthe estimates obtained; it may also be possible to discriminate between additive and multiplicative covariate effects.nConclusions: Generalised linear models are attractive for the regression of cost data because they providenparametric methods of analysis where a variety of non-normal distributions can be specified and the way covariatesnact can be altered. Unlike the use of data transformation in ordinary least-squares regression, generalised linearnmodels make inferences about the mean cost directly.
机译:目的:选择一种合适的方法对成本数据进行回归分析是有问题的,因为它必须关注人口均值,同时要考虑到数据的典型偏斜分布。在本文中,wenillustrate使用广义线性模型对成本数据进行回归分析。n方法:我们考虑具有身份链接函数(提供加性协变量效应)或对数链接函数(提供乘法效应)和高斯(正态分布)的广义线性模型。 ),泊松,纳马或高斯逆分布过度分散。这些被用于估计在针对基线协变量进行调整的两个随机试验中的治疗效果。结果:在两个示例中,高斯模型拟合性较差,其他分布是优选的。n当模型中具有预后重要性的变量时,使用不同的分布会严重影响获得的估计值。结论:广义线性模型对成本数据的回归很有吸引力,因为它们提供了可以指定各种非正态分布以及可以指定协变量的方式的分析方法。改变了。与普通最小二乘回归中使用数据转换不同,广义线性模型直接推断平均成本。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号