str(t1)
'data.frame':   45846 obs. of  3 variables:
$ year: int  2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 ...
$ Bdcd: int  112001 112001 112001 112001 112001 112001 112001 112001 112001 112001 ...
$ age : num  0.589 0.597 0.6 0.603 0.605 0.608 0.616 0.619 0.622 0.625 ...
> temp1<-melt(t1,id=c('year','Bdcd'))
> temp2<-cast(temp1,year+Bdcd~variable,mean,na.rm=T)
> str(temp2)
List of 3
$ year: int [1:618] 2007 2007 2008 2008 2008 2008 2008 2008 2008 2008 ...
$ Bdcd: int [1:618] 122000 122001 112001 112002 112003 112004 112005 112006 122000 122001 ...
$ age : num [1:618] 0.1569 0.0807 0.4268 0.3747 0.227 ...
- attr(*, "row.names")= int [1:618] 1 2 3 4 5 6 7 8 9 10 ...
- attr(*, "idvars")= chr [1:2] "year" "Bdcd"
- attr(*, "rdimnames")=List of 2
  ..$ :'data.frame':    618 obs. of  2 variables:
  .. ..$ year: int [1:618] 2007 2007 2008 2008 2008 2008 2008 2008 2008 2008 ...
  .. ..$ Bdcd: int [1:618] 122000 122001 112001 112002 112003 112004 112005 112006 122000 122001 ...
  ..$ :'data.frame':    1 obs. of  1 variable:
  .. ..$ variable: Factor w/ 1 level "age": 1
> fix(temp2)
> str(temp2)
'data.frame':   618 obs. of  3 variables:
$ year: num  2007 2007 2008 2008 2008 ...
$ Bdcd: num  122000 122001 112001 112002 112003 ...
$ age : num  0.1569 0.0807 0.4268 0.3747 0.227 ...
为什么fix前后temp2的结构会发生变化，cast函数有没有错？