1

我正在尝试将模型拟合到我的rating(1,2,3) 和group(1,2,3) 的有序分类数据中。其他预测因子是age(18-23, 24-30),continent和medication(y/n) 的使用。我一直在使用有用的教程：https ://journals.sagepub.com/doi/full/10.1177/2515245918823199来指导我完成我的第一个贝叶斯回归，但我希望有人可以帮助解释如何“检查”模型？

我运行了以下模型：

modelall<- brm(formula = rating ~ 1 + cs(group) + medication + continent +agequartiles,
data = df,
family = acat(“cloglog”))

modelagroup<- brm(formula = rating  ~ 1 + cs(group),
data = df,
family = acat("cloglog"))

modelage<- brm(formula = rating  ~ 1 + cs(group)  + agequartiles,
data = df,
family = acat("cloglog"))

modelmeds<- brm(formula = rating  ~ 1 + cs(group)  + medication,
data = df,
family = acat("cloglog"))

我已经习惯了使用 P 值我实际上不确定如何解释我的输出

> summary(model group)
 Family: acat 
  Links: mu = cloglog; disc = identity 
Formula: rating ~ 1 + cs(group) 
   Data: df (Number of observations: 2637) 
Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
         total post-warmup samples = 4000

Population-Level Effects: 
               Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept[1]       2.81      0.17     2.49     3.14 1.01      697     1156
Intercept[2]      -0.56      0.09    -0.73    -0.40 1.01      682     1341
group2[1]          0.81      0.20     0.42     1.20 1.01      765     1399
group2[2]         -0.47      0.12    -0.70    -0.24 1.01      855     1508
group3[1]          0.83      0.20     0.45     1.23 1.01      684     1328
group3[2]         -0.45      0.12    -0.68    -0.22 1.01      805     1533

Family Specific Parameters: 
     Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
disc     1.00      0.00     1.00     1.00 1.00     4000     4000

Samples were drawn using sampling(NUTS). For each parameter, Bulk_ESS
and Tail_ESS are effective sample size measures, and Rhat is the potential
scale reduction factor on split chains (at convergence, Rhat = 1).

如果有人可以提供帮助，我将不胜感激，根据我的理解应该在 1.1 和 ESS 尽可能高。由于不是我样本中的每个人都同意回答每个问题，我似乎无法做一个 LOO，因为他们有不同数量的观察。除了查看 CI 是否超过 0 之外，我只是不确定如何计算模型中应该包含的内容。如何检查我的模型是否正常工作？

谢谢

4

0 回答 0