gsi.recode {compositions} | R Documentation |
The missing codes can be replace with arbitrary value e.g. IEEE numbers and viceversa through this interface.
gsi.recodeM2C(x,y=x,BDL,SZ,MAR,MNAR,NMV) gsi.recodeC2M(x,y=x,na,nan,ninf,inf,neg,zero,pos) gsi.recodeM2Clean(x,y=x,BDL=NaN,SZ=NaN,MAR=NaN,MNAR=NA,NMV) gsi.cleanR(x)
x |
the dataset having missings or IEEE numbers |
y |
a dataset of similar shape, where the replacment should take place |
BDL |
value to replace for BDL |
SZ |
value to replace for SZ |
MAR |
value to replace for MAR |
MNAR |
value to replace for MNAR |
NMV |
value to replace for NMV |
na |
value to replace for NA |
nan |
value to replace for NaN |
ninf |
value to replace for -Inf |
inf |
value to replace for Inf |
neg |
value to replace for numbers with x<0 |
zero |
value to replace for numbers with x==0 |
pos |
value to replace for numbers with x>0 |
y with entries replaced. gsi.cleanR replaces all improper numbers with 0.
Do not use gsi.* functions directly since they are internal functions of the package
K.Gerald v.d. Boogaart http://www.stat.boogaart.de
gsi.plain(acomp(c(12,3,4)))