See also
1. official command -inbase-.
2. -egen- function -base()- in -egenmore- from SSC. The help gives an
example that seems parallel to Steinar's problem.
Note that -split- (as suggested by Martin Weiss) is no help here.
-split- is for strings subdivided by parsing or punctuation characters.
Nick
[email protected]
Kit Baum
Given an integer between 0-255 and the desired bit to be retrieved
(counting from the left or high-order bit), this routine will give you
the binary form of the integer and the bit as locals:
mata: mata clear
mata:
void bit(real scalar v, real scalar w)
{
string vector d, b
d=("0000","0001","0010","0011","0100","0101","0110","0111", //
"1000","1001","1010","1011","1100","1101","1110","1111")
for(i=1;i<=cols(d);i++) {
dd = d[i] :+ d
b = b, dd
}
if( v>0 & v <= 255 & w > 0 & w <= 8) {
st_local("v" , b[v+1] )
st_local("w", substr(b[v+1], w, 1))
}
}
end
. mata: bit(128,1)
. di "`v' `w'"
10000000 1
. mata: bit(128,3)
. di "`v' `w'"
10000000 0
Steinar Fossedal
================
I have imported an 8-bit integer variable where each bit represents a
flag telling me if some property of the observation is true or false.
E.g. the number 8 in binary format is 00001000, while 10 would be
00001010. Bit nr. 4 from the left tells me if a car is delivered or not,
while the second to the right tells me if it's paid or not. So 8 = car
is delivered but not paid, 2 = paid but not delivered, 10 = paid and
delivered, etc.
I need to split this information into 8 dummyvariables so they can be
used to segment my data.
Does there exist a fairly smooth way to do this in Stata?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
