我在数据集上预算小鼠：
lwp.imputex＆lt; - 小鼠（databasename，m = 5，maxit = 50，seed = 500）
之后，我在此之后我在汇总的数据集上进行了线性回归：
pbmi_modelx＆lt; - 带有（数据= lwp.Imputex，exp = lm（sds_bmi_y5_iotf〜paternal_reported_bmi_bl））摘要（pbmi_modelx）
ATED此分析的摘要统计数据当我尝试获得汇总数据集的95％置信区间时，我会收到以下错误消息 -
confint（pool_pbmi_modelx）
t.test.default中的错误（x，y = y，替代=替代方案，mu = mu，：
没有足够的“ x”观察结果，
我不确定为什么置信区间不像我那样使用“摘要”或“ confint”命令对我有用，就像在我的竞争案例分析中一样，我可以以这种方式生成它们。
任何帮助都将不胜感激，或者如果有人可以为来自非统计背景的人推荐的有关归纳小鼠的非统计背景的人提供任何好的参考/来源。谢谢。
I preformed mice imputation on my dataset:
LWP.imputeX<- mice(databasename, m=5, maxit=50, seed=500)
Following this I ran a linear regression on the pooled dataset:
PBMI_ModelX <- with(data=LWP.imputeX, exp = lm(SDS_BMI_y5_IOTF ~ Paternal_reported_bmi_BL)) summary(PBMI_ModelX) Pool_PBMI_ModelX<-pool(PBMI_ModelX) summary(Pool_PBMI_ModelX)
Although I generated the summary statistics for this analysis when I try to get the 95% confidence intervals for the pooled dataset I receive the following error message -
confint(Pool_PBMI_ModelX)
Error in t.test.default(x, y = y, alternative = alternative, mu = mu, :
not enough 'x' observations
I'm not sure why the confidence intervals won't work for me using the 'summary' or 'confint' commands as in my compete cases analysis I can generate them in this way.
Any assistance would be appreciated or if anyone could recommend any good references/sources for people from a non stats background on running MICE imputation. Thanks.
发布评论