用R中的gamlssMX()函数做有限混合伽马回归时遇到错误

代码及结果如下
> Mix1<-gamlssMX(logcl~veh_body+vvalueg+gender+area+agecat,data=dats,family=GA,K=2)
Error in if (dv > olddv && itn >= 2 && auto == TRUE) { :
  missing value where TRUE/FALSE needed
In addition: Warning message:
In glim.fit(f = mu.object, X = mu.X, y = y, w = w, fv = mu, os = mu.offset,  :
  The deviance has increased in an inner iteration for mu
Increase gd.tol and if persist, try different steps
or model maybe inappropriate

我试了一下，删去agecat变量后可以进行拟合，原始数据有4712条记录，会不会是分类太多导致迭代的时候数据不够用啊？（瞎猜的），请各位大神赐教

求助啊

在线等，写论文遇到瓶颈啊