受symnum()函数启发,写了一个缺失数据可视化函数(类似VIM包的作图方法):

<br />
# shrink表示是否将缺失部位相同的行合并<br />
# rownames表示是否使用行名<br />
# abbr表示是否简化列名<br />
na.mat=function(z,reorder=T,shrink=T,rownames=T,abbr=F){<br />
  if(is.null(rownames(z)))rownames(z)=1:nrow(z)<br />
  as.data.frame(z)->z->y<br />
  sapply(z,is.na)->z<br />
  rowsum=apply(z,1,sum)<br />
  colsum=apply(z,2,sum)<br />
  z[rowsum>0,colsum>0]->z<br />
  rownames(z)=rownames(y)[rowsum>0]<br />
  if(reorder){<br />
    z[order(rowsum[rowsum>0]),order(colsum[colsum>0])]->z<br />
  }<br />
  insert=function(x,what,where){<br />
    what=what[rep(1:length(what),len=length(where))]<br />
    what=what[order(where)]<br />
    where=sort(where)<br />
    numeric(length(x)+length(where))->z<br />
    where1=where+1:length(where)<br />
    z[where1]=what<br />
    z[-where1]=x<br />
    unlist(z)<br />
  }<br />
  pas=function(z){<br />
    c(as.matrix(z))->z<br />
    if(length(z)>1){</p>
<p>      z[-1]->v<br />
      as.numeric(z)->z<br />
      c(0,diff(z))->dif<br />
      by(z,cumsum(dif!=1),function(t){<br />
        if(length(t)>1)paste(head(t,1),tail(t,1),sep='~')->t<br />
        t<br />
      })->out<br />
      paste(unlist(out),collapse=' ')->z<br />
    }<br />
    as.character(z)<br />
  }<br />
  if(shrink){<br />
    as.data.frame(z)->x->q<br />
    unique(x)->x<br />
    by(rownames(z),cumsum(!duplicated(q)),pas)->rownames(x)<br />
    as.matrix(x)->z<br />
  }<br />
  if(reorder){<br />
    z[order(rowSums(z)),order(colSums(z))]->z<br />
  }<br />
  if(!rownames)rownames(z)=1:dim(z)[1]<br />
  cat('nrow:',nrow(y),', ncol:',ncol(y),'\n')<br />
  print(symnum(z,cut=c(0,.5,1),symbol=c('.','*'),abbr=abbr))<br />
  invisible(z)<br />
}<br />

拿DMwR包的数据algae来试一下:

<br />
> na.mat(algae)<br />
nrow: 200 , ncol: 18<br />
        mxPH mnO2 NO3 NH4 oPO4 PO4 Cl Chla<br />
28      .    .    .   .   .    *   .  .<br />
38      .    *    .   .   .    .   .  .<br />
48      *    .    .   .   .    .   .  .<br />
63 116  .    .    .   .   .    .   .  *<br />
161 184 .    .    .   .   .    .   *  .<br />
55~61   .    .    .   .   .    .   *  *<br />
62      .    *    *   *   *    .   *  *<br />
199     .    .    *   *   *    *   *  *<br />
> na.mat(algae,abbr=T)<br />
nrow: 200 , ncol: 18<br />
        mP mO NO NH o P Cl Ch<br />
28      .  .  .  .  . * .  .<br />
38      .  *  .  .  . . .  .<br />
48      *  .  .  .  . . .  .<br />
63 116  .  .  .  .  . . .  *<br />
161 184 .  .  .  .  . . *  .<br />
55~61   .  .  .  .  . . *  *<br />
62      .  *  *  *  * . *  *<br />
199     .  .  *  *  * * *  *<br />
> na.mat(algae,abbr=T,shrink=F)<br />
nrow: 200 , ncol: 18<br />
    mP mO NO NH o P Cl Ch<br />
28  .  .  .  .  . * .  .<br />
38  .  *  .  .  . . .  .<br />
48  *  .  .  .  . . .  .<br />
63  .  .  .  .  . . .  *<br />
116 .  .  .  .  . . .  *<br />
161 .  .  .  .  . . *  .<br />
184 .  .  .  .  . . .  *<br />
55  .  .  .  .  . . *  *<br />
56  .  .  .  .  . . *  *<br />
57  .  .  .  .  . . *  *<br />
58  .  .  .  .  . . *  *<br />
59  .  .  .  .  . . *  *<br />
60  .  .  .  .  . . *  *<br />
61  .  .  .  .  . . *  *<br />
62  .  *  *  *  * . *  *<br />
199 .  .  *  *  * * *  *<br />

</p>