* Syntax creates AFCARS Foster Care Annual File for 2009 . 
* from converted submissions 2009 A and 2009 B.

GET
  FILE='c:\composite\xxx10a.sav'.
compute priorbeg=yrmoda(2009,09,30).
compute periodbeg=yrmoda(2009,10,01).
compute periodend=yrmoda(2010,09,30).
compute afterend=yrmoda(2010,10,01).
compute entryday=YRMODA(latremyr,latremmo,latremda).
compute exitday=YRMODA(dodfcyr,dodfcmo,dodfcda).
compute dob=yrmoda(dobyr,dobmo,dobda).
compute key=1.
formats key (f1.0).
exe.
save outfile='c:\composite\tempxxx10a.sav'.

GET
  FILE='c:\composite\xxx10b.sav'.
compute priorbeg=yrmoda(2009,09,30).
compute periodbeg=yrmoda(2009,10,01).
compute periodend=yrmoda(2010,09,30).
compute afterend=yrmoda(2010,10,01).
compute entryday=YRMODA(latremyr,latremmo,latremda).
compute exitday=YRMODA(dodfcyr,dodfcmo,dodfcda).
compute dob=yrmoda(dobyr,dobmo,dobda).
compute key=2.
formats key (f1.0).
exe.
save outfile= 'c:\composite\tempxxx10b.sav'.

get file= 'c:\composite\tempxxx10a.sav'.
ADD FILES /FILE=*
 /FILE='c:\composite\tempxxx10b.sav'.
exe.

/** Identify Duplicate Cases - Using Record Number and FIPS.

SORT CASES BY recnumbr(A) fipscode(A) key(A) .
MATCH FILES /FILE = * /BY recnumbr fipscode 
 /FIRST = PrimaryFirst /LAST = PrimaryLast.
DO IF (PrimaryFirst).
COMPUTE MatchSequence = 1 - PrimaryLast.
ELSE.
COMPUTE MatchSequence = MatchSequence + 1.
END IF.
LEAVE MatchSequence.
FORMAT MatchSequence (f7).
COMPUTE InDupGrp = MatchSequence > 0.
SORT CASES InDupGrp(D).
MATCH FILES /FILE = * /DROP = PrimaryFirst InDupGrp.
VARIABLE LABELS PrimaryLast 'Indicator of each last matching case as Primary'
  MatchSequence 'Sequential count of matching cases' .
VALUE LABELS PrimaryLast 0 'Duplicate Case' 1 'Primary Case'.
VARIABLE LEVEL PrimaryLast (ORDINAL)
     /MatchSequence (SCALE).
FREQUENCIES VARIABLES = PrimaryLast MatchSequence .
EXECUTE.
filter off.
use all.
select if(primarylast=1).
exe.

/** Identify Duplicate Cases - Using Record Number and DOB.

SORT CASES BY recnumbr(A) dob(A) key(A) .
MATCH FILES /FILE = * /BY recnumbr dob
 /DROP = PrimaryLast MatchSequence /FIRST = PrimaryFirst /LAST = PrimaryLast.
DO IF (PrimaryFirst).
COMPUTE MatchSequence = 1 - PrimaryLast.
ELSE.
COMPUTE MatchSequence = MatchSequence + 1.
END IF.
LEAVE MatchSequence.
FORMAT MatchSequence (f7).
COMPUTE InDupGrp = MatchSequence > 0.
SORT CASES InDupGrp(D).
MATCH FILES /FILE = * /DROP = PrimaryFirst InDupGrp.
VARIABLE LABELS PrimaryLast 'Indicator of each last matching case as Primary'
  MatchSequence 'Sequential count of matching cases' .
VALUE LABELS PrimaryLast 0 'Duplicate Case' 1 'Primary Case'.
VARIABLE LEVEL PrimaryLast (ORDINAL)
     /MatchSequence (SCALE).
FREQUENCIES VARIABLES = PrimaryLast MatchSequence .
EXECUTE.
filter off.
use all.
select if (primarylast=1).
exe.
COMPUTE NStatus=1.
exe.
IF (MISSING(entryday) and (exitday > 0)) NStatus=2.
exe.
value labels NStatus
 1'NExists'
 2'Missing N'.
exe.
freq NStatus.

*review output and determine whether there are a large proportion of cases with missing entry dates.

compute flag=0.
IF ((MISSING(latremyr) and (YRMODA(dodfcyr,dodfcmo,dodfcda)) <= periodend and (yrmoda(dodfcyr,dodfcmo,dodfcda)) >=
 periodbeg or (yrmoda(latremyr,latremmo,latremda)) <= periodend and (missing(dodfcyr)) or (yrmoda(dodfcyr,dodfcmo,dodfcda)) >= periodbeg)) 
 flag=1.
freq flag.

*review output and determine whether there are a large proportion of cases where flag=0; if so, this means that there are cases that were 
*outside of the reporting period.

sel if flag=1. 

exe.
SAVE OUTFILE='c:\composite\xxx10ab.sav'
  /DROP=flag NStatus priorbeg periodbeg periodend afterend entryday exitday dob key PrimaryLast MatchSequence.