lmer 中使用的 b 样条出现错误“derivs 大于 x 的长度”

Question

我用

Rstudio

。根据https://stats.stackexchange.com/questions/135694/using-b-splines-within-a-linear-mixed-effects-model-in-r我想添加所谓的b样条线进入我的 lmer 模型：

degree<- 3
fit<- lmer(log(log(Amplification)) ~ bs(poly(Voltage, degree)) + bs(poly(Voltage, degree) | Serial_number),data = APD))

但这会导致

Error in splineDesign(Aknots, x, ord) : 
  length of 'derivs' is larger than length of 'x'
In addition: Warning messages:
1: In Ops.factor(poly(Voltage, 3), Serial_number) :
  ‘+’ not meaningful for factors
2: In min(x, na.rm = na.rm) :
  no non-missing arguments to min; returning Inf
3: In max(x, na.rm = na.rm) :
  no non-missing arguments to max; returning -Inf

这是什么意思？当我使用时

summary(fit<- lmer(log(log(Amplification)) ~ (poly(bs(Voltage), degree)) + (poly(Voltage, degree) | Serial_number), data = APD))

然后这会导致

Error in poly(dots[[i]], degree, raw = raw, simple = raw) : 
  'degree' must be less than number of unique points

下降到

degree=0

收益率

Error in poly(dots[[1L]], degree, raw = raw, simple = raw && nd > 1) : 
  'degree' must be at least 1

如何将 B 样条线添加到 lmer 中？ 预先感谢您！

Answer 1

在回归中使用尺度（VAR）后，这发生在我身上。当我在上一步中使用 mutate(CVAR=scale(VAR, center=TRUE, scale=TRUE) 缩放变量时，ggpredict 给出了正确的估计。