作者cropcircles (神秘的麦田圈)
看板Statistics
标题[程式] SAS logistic regression
时间Wed Apr 4 11:13:47 2012
[软体程式类别]:
SAS
[程式问题]:
logistic regression的资料处理
[软体熟悉度]:
低(1~3个月)
[问题叙述]:
我有一个a档： 资料笔数有6500多笔，变数有13个、
dummy变数有26个(产业别21个、年度5个)，这个档里没有任何missing
问题：但是proc means时，最後面的7个dummy variable皆显示为N=0 mean=.
读取不到任何资料，请问是否是因为变数太多资料太多造成的?
NOTE: 6533 records were read from the infile 'H:\0321\output\datamerge.txt'.
The minimum record length was 102.
The maximum record length was 118.
One or more lines were truncated.
NOTE: The data set WORK.A has 6533 observations and 39 variables.
NOTE: DATA statement used (Total process time):
real time 0.03 seconds
cpu time 0.01 seconds
和b、c档merge之後，继续跑logistic regeression
ERROR: There are no valid observations.
NOTE: The SAS System stopped processing this step because of errors.
NOTE: There were 6241 observations read from the data set WORK.TOTAL1.
NOTE: PROCEDURE LOGISTIC used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds
我找了好久也有把error的叙述去搜寻了，可是一直找不到原因
请大家帮忙一下，谢谢!
[程式范例]:
data a;
infile 'H:\0321\output\datamerge.txt' expandtabs firstobs=2 truncover;
input CODE LNTA GROWTH ROA LEV ARINV YEAR CLISTB CLISTH CLISTF OPINION INDE
OWN ind2 ind3 ind4 ind5 ind6 ind7 ind8 ind9 ind10 ind11 ind12
ind13 ind14 ind15 ind16 ind17 ind18 ind19 ind20 ind21 ind22
yr2 yr3 yr4 yr5 yr6;
proc sort; by CODE YEAR ;
proc means ;
.
.
.
.(中间是merge的过程，这部分已经确认没有错误)
.
.
.
.
data total1;set total;
proc logistic descending ;
model BIG4=PRETRADE LNTA GROWTH ROA LEV ARINV CLISTB CLISTH CLISTF OPINION
INDE OWN ind2 ind3 ind4 ind5 ind6 ind7 ind8 ind9 ind10 ind11
ind12 ind13 ind14 ind15 ind16 ind17 ind18 ind19 ind20 ind21 ind22
yr2 yr3 yr4 yr5 yr6;
run;
--
※ 发信站: 批踢踢实业坊(ptt.cc)
◆ From: 122.122.220.188