小鼠中的 complete 究竟是做什么的? [英] What exactly does complete in mice do?

问题描述

我正在研究如何使用多重插补结果.以下是我的理解，如有错误请指出.

I am researching how to use multiple imputation results. The following is my understanding, and please let me know if there're mistakes.

假设您有一个包含缺失值的数据集，并且您想要进行回归分析.您可以对 m = 5 次执行多重插补，并且对于每个插补数据集(现在是 5 个插补数据集)，您运行回归分析，然后池"执行回归分析.通过鲁宾规则(或使用 R 包pool")从这些 m = 5 模型中估计系数.

Suppose you have a data set with missing values, and you want to conduct a regression analysis. You may perform multiple imputation for m = 5 times, and for each imputed data set (5 imputed data sets now) you run a regression analysis, then "pool" the coefficient estimates from these m = 5 models via Rubin's rules (or use R package "pool").

我的问题是，在鼠标中你有一个函数complete()，手册说你可以使用complete(object)提取完整的数据集.

My question is that, in mice you have a function complete(), and the manual says you can extract completed data set by using complete(object).

但是如果我使用鼠标 m = 5 次，使用 complete() 仍然有意义吗?complete() 将为我得到哪些插补结果?

But if I use mice for m = 5 times, does it still make sense to use complete()? Which imputation results will complete() get for me?

另外，如果我只使用 m = 1 的鼠标有意义吗?谢谢.

Also, does it make sense if I only use mice with m = 1? Thank you.

推荐答案

您可能忽略了 mice::complete() 在参数中使用 action=1 作为默认值，其中返回第一个估算数据集"(参见?mice::complete)并且实际上毫无价值.

You probably overlooked that mice::complete() in arguments uses action=1 as default, which "returns the first imputed data set" (see ?mice::complete) and actually is worthless.

您绝对应该使用 action="long" 来考虑多重性"多重插补！

You should definitely use action="long" to take account for the "multiplicity" of the multiple imputation!

不，使用 m=1 完全没有意义(除了调试)，因为每个插补都是基于一个随机过程，你必须汇集结果(使用任何方法)来解释变化.通常推荐 m>20¹.

No, it makes no sense at all to use m=1 (apart from debugging), because every imputation is based on a random process and you have to pool the results (using any method whatsoever) to account for the variation. Often m>20 is recommended¹.

基本上，多重插补的工作原理如下:

Basically, multiple imputation works as follows:

1. 用一个随机分量创建 m 个插补过程，以获得
2. m 个略有不同的估算数据集.
3. 分析每个插补数据集以获得略有不同的参数估计.
4. 组合结果，计算参数估计值的变化.

1. Create m imputation processes with a random component, to obtain
2. m slightly different imputed data sets.
3. Analyze each imputed data set to get slightly different parameter estimates.
4. Combine results, calculating the variation in parameter estimates.

(另见 multiple-imputation-in-a-nutshell 简要概述.)

(Also see multiple-imputation-in-a-nutshell for a brief overview.)

这篇关于小鼠中的 complete 究竟是做什么的?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！