Question

我正在研究Pinhiero和Bates在R中的S和S-Plus中的混合效果模型。我很难使第8章（p。387）中的模型收敛。 / p>

library(nlme)

fm1Wafer.nlmeR <- nlme(current ~ A + B * cos(4.5679 * voltage) + C * sin(4.5679 * voltage),
                       data = Wafer,
                       fixed = list(A ~ voltage + I(voltage^2), B + C ~ 1),
                       random = list(Wafer = A ~ voltage + I(voltage^2),
                                     Site = pdBlocked(list(A ~ 1, A ~ voltage + I(voltage^2)-1))),
                       start = c(-4.26, 5.62, 1.26, -0.10, 0.10), # starting values taken from fixed effects of another model earlier in the book
                       method = "REML",
                       control = nlmeControl(opt = "nlm"))

如您所见，我已经尝试过nlm优化器。默认的nlminb优化器也不起作用。两者都会产生此错误消息

Error in nlme.formula(current ~ A + B * cos(4.5679 * voltage) + C * sin(4.5679 *  : 
  maximum number of iterations (maxIter = 50) reached without convergence
In addition: Warning messages:
1: In logLik.reStruct(object, conLin) :
  Singular precision matrix in level -2, block 1
2: In logLik.reStruct(object, conLin) :
  Singular precision matrix in level -2, block 1

有什么建议吗？本书的下游有几种基于该模型的模型，因此最好使其收敛。

Answer 1

该错误表明设置nlmeControl(opt = "nlm", maxIter = 2000)之类的内容会有所帮助，但实际上没有帮助。我尝试了4000次，但时间长了...

真正的罪魁祸首是method = "REML"。保留默认的"ML"可获得预期的结果。

fm1Wafer.nlmeR <- nlme(current ~ A + B * cos(4.5679 * voltage) + C * sin(4.5679 * voltage),
                       data = Wafer,
                       fixed = list(A ~ voltage + I(voltage^2), B + C ~ 1),
                       random = list(Wafer = A ~ voltage + I(voltage^2),
                                     Site = pdBlocked(list(A ~ 1, A ~ voltage + I(voltage^2) - 1))),
                       start = c(-4.26, 5.62, 1.26, -0.10, 0.10))

在Pinhiero和Bates的书中，非线性混合效应模型无法收敛的问题

1 个答案: