SPSS23.0做倾向性匹配缺FUZZY

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

用spss23.0做PSM的时候，输出的结果如图，请问各位好汉，怎么解决呢？* Propensity Score Matching.
begin program.
import spss, random, spssaux, codecs
try:
   import FUZZY
except:
   print "This procedure requires the FUZZY extension command which is not installed."
parts = [int(i) for i in FUZZY.__version__.split(".")]
if  parts < [1,3,0]:
   raise EnvironmentError("This procedure requires version 1.3.0 or later of FUZZY")
ds = spss.ActiveDataset()
if ds =="*":
   ds = "D" + str(random.random())
   spss.Submit("DATASET NAME " + ds)
ds = codecs.encode(ds)  # must use code page rep here
matchgroup = "M" + str(random.random())  #temporary variable
tempdsname = "D" + str(random.random())
lrcmd = spssaux.u(r"""LOGISTIC REGRESSION VARIABLES Objectgroup
     /METHOD=ENTER Age Eduction Marriage Children
     /SAVE=PRED(PS).""")
try:
   spss.Submit(lrcmd)
except:
   raise ValueError("Logistic Regression step failed")
fuzzycmd = spssaux.u(r"""FUZZY BY=PS SUPPLIERID=ID NEWDEMANDERIDVARS=AA
GROUP=Objectgroup EXACTPRIORITY=FALSE
DRAWPOOLSIZE=CC MATCHGROUPVAR=%(matchgroup)s FUZZ=0.1 DS3=%(tempdsname)s
/OPTIONS SAMPLEWITHREPLACEMENT=FALSE MINIMIZEMEMORY=TRUE SHUFFLE=FALSE .
DELETE VARIABLES %(matchgroup)s.
DATASET ACTIVATE %(ds)s.
DELETE VARIABLES %(matchgroup)s.
DATASET COPY BB.
DATASET ACTIVATE BB.
SELECT IF Objectgroup EQ 1.
DATASET ACTIVATE %(tempdsname)s.
DATASET ACTIVATE BB.
ADD FILES /FILE=* /FILE=%(tempdsname)s.
EXECUTE.
DATASET CLOSE %(tempdsname)s.""" % locals())
spss.Submit(fuzzycmd)
end program.
FUZZY BY=Age Eduction Marriage Children SUPPLIERID=ID NEWDEMANDERIDVARS=D GROUP=Objectgroup
    FUZZ=0.1 EXACTPRIORITY=FALSE
DRAWPOOLSIZE=F MATCHGROUPVAR=E
/OPTIONS SAMPLEWITHREPLACEMENT=FALSE MINIMIZEMEMORY=TRUE SHUFFLE=FALSE.

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Fuzzy SPSS 倾向性 PSS replacement

自己回一个，还是插件的问题，建议要做PSM的小伙伴用SPSS26.0版本靠谱

https://bbs.pinggu.org/thread-11212308-1-1.html倾向得分匹配问题可以查看这篇文章，SPSS插件太麻烦的话，可以使用SPSSAU比较方便简单