st: Patterns of longitudinal data

[Dick Campbell <dcamp@uic.edu>]

Guillermo Cruces said:

I am working with panel data and I found the program xtpattern extremely 
useful.
<snip>
What I need now is to know how many contiguous observations there are, so for
instance:
1.1..1 would be zero
111111 would be six
111.11  would be (perhaps) three (the max).

The obvious thing would be to parse through the string generated by 
xtpattern...
but I couldn't find an elegant way around this p

xtpattern generates a string variable in the form 1.11.1 etc.
This can be interpreted as a binary number. Suppose there
are three observations maximum and we create the decimal
equivalent of the binary numbers resulting from xtpattern.
The fist position expands as 2**2, the second as 2**1
and the third as 2**0  (= 1). Thus for each pattern of  observations
there is a unique decimal number,
Here is an example for four possible patterns.
111 = 7
11. = 6
1.1 = 5
1.. = 4

A neat characteristic of the decimal number is that
the higher the number of consecutive observations at the beginning of the
series the higher the decimal number. The second observation has a higher 
number than the third.
etc.  You will also see that "consecutive observations always generate 
higher numbers
that non-consecutive, regardless of pattern.

=====================================
Richard T. Campbell
Professor of Sociology
Phone:          312/413-3759
Fax:                    312/996-5104
More info:              http://www.uic.edu/~dcamp

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/