考虑下面的纵向/重复测量数据集示例
library(tidyverse)
library(broom)
library(nlme)
data <- read.csv("https://stats.idre.ucla.edu/stat/data/study2.csv")
data <- data %>% mutate(dbp =rnorm(120, 30:150), sbp = rnorm(120, 50:200),bmi
= rnorm(120,15:40), chol = rnorm(120,50:350), insulin = rnorm(120,2:40), educ = rnorm(120,5:10))
我可以用
group_by %>% do(tidy(*))
运行几个未经调整和调整的
单水平
回归模型(通过结果和风险列表循环)并将模型结果提取到数据框架中,如下所示
out <-c("pulse","insulin","chol")
exp <- c("factor(exertype)","sbp","dbp")
conf <- c("bmi","factor(diet)")
#Unadjusted models - single level regression (lm)
#################################################
Unadjusted <- expand.grid(out, exp) %>%
group_by(Var1) %>% rowwise() %>%
summarise(frm = paste0(Var1, "~", Var2)) %>%
group_by(model_id = row_number(),frm) %>%
do(tidy(lm(.$frm, data = data))) %>%
mutate(lci = estimate-(1.96*std.error)) %>%
mutate(uci = estimate+(1.96*std.error))
#Adjusted models - single level regression (lm)
###############################################
Adjusted <- expand.grid(out, exp, conf) %>%
group_by(Var1, Var2) %>%
summarise(Var3 = paste0(Var3, collapse = "+")) %>%
rowwise() %>%
summarise(frm = paste0(Var1, "~", Var2, "+", Var3)) %>%
group_by(model_id = row_number(), frm) %>%
do(tidy(lm(.$frm, data = data))) %>%
mutate(lci = estimate-(1.96*std.error)) %>%
mutate(uci = estimate+(1.96*std.error))
我想用同样的方法来适应
多层次
用于解释重复数据的模型。使用示例代码:
lme(sbp ~ pulse+factor(diet)+time, data=data, random= ~time|id, method ="ML")
但是,当我尝试这样做时,例如使用:
#Unadjusted models - multi-level regression (lme)
#################################################
Unadjusted <- expand.grid(out, exp) %>%
group_by(Var1) %>% rowwise() %>%
summarise(frm = paste0(Var1, "~", Var2)) %>%
group_by(model_id = row_number(),frm) %>%
do(tidy(lme(.$frm, data = data, random= ~time|id, method = "ML"))) %>%
mutate(lci = estimate-(1.96*std.error)) %>%
mutate(uci = estimate+(1.96*std.error))
我收到以下错误消息:
Error in UseMethod("lme") : no applicable method for 'lme' applied to an object of class "character"
有什么想法,如何使这个工作的lme型模型?
