a********6
发帖数: 14468 | 1 There are hundreds of variables in my csv file. I tried to let SAS read the
first 30 rows as variables (they are indeed!) and SAS refused to do so.
Please help me this out. Crazy job.
Thanks! |
w*****e
发帖数: 806 | |
a********6
发帖数: 14468 | 3 Thank you. I looked through the page and it does not seem to have mentioned
my case: too many variables and they have to occupy many rows in the txt or
csv format.
【在 w*****e 的大作中提到】
: proc import ??
: http://www.ats.ucla.edu/stat/sas/faq/read_delim.htm
|
l*********s
发帖数: 5409 | |
A****t
发帖数: 141 | 5 1000?
data one;
infile "...." dlm=',' lrecl=100000;
input var1-var1000;
run; |
a********6
发帖数: 14468 | 6 这个正是想问的问题，SAS 9的极限是多少个变量？
我几百个变量的名字一定要保留，因为不想再手工输入，所以着急啊。
【在 A****t 的大作中提到】
: 1000?
: data one;
: infile "...." dlm=',' lrecl=100000;
: input var1-var1000;
: run;
|
h***i
发帖数: 634 | 7 "Beginning in SAS 9, the maximum number of variables allowed is no longer 32
,767."
"You can define a data set with an observation length of up to 2 GB on a 32-
bit platform and approximately 2**46 on a 64-bit platform."
check http://support.sas.com/kb/8/213.html |
A****t
发帖数: 141 | |
a********6
发帖数: 14468 | 9 我是新新手，所以还是用SPSS迅速搞定了。SAS应该也可以，但不自己手工输入变量可
能不行。
【在 A****t 的大作中提到】
: http://support.sas.com/kb/8/213.html
: proc import不能读吗？
|
a********6
发帖数: 14468 | 10 终于在SAS里实现了，csv和excel都可以import,只是变量多，懒得把它们的属性一一修
改，不知道在分析数据时会有什么危害，欢迎懂行的人给三言两语科普一下。 |
D******n
发帖数: 2836 | 11 you mean u just go with the format SAS automatically assigns? thats not
very
good. In industry, the raw data should always come with a layout, u just
use that layout to read in your raw data. If not so, you have to look into
each field to figure it out. SAS gets it mostly right .
I guess u might be wrong about your statement in your first post too.
usually one row per record as well as for the header too. It might look
like
multiple rows in your text editor but it is technically just one row. |
a********6
发帖数: 14468 | 12 "It might look like multiple rows in your text editor but it is technically
just one row."
是啊，你的意思是CSV文件其实可以有记忆，将看起来多行的东西记为一行？而且，SAS
可以识别这种记忆？谢谢。
【在 D******n 的大作中提到】
: you mean u just go with the format SAS automatically assigns? thats not
: very
: good. In industry, the raw data should always come with a layout, u just
: use that layout to read in your raw data. If not so, you have to look into
: each field to figure it out. SAS gets it mostly right .
: I guess u might be wrong about your statement in your first post too.
: usually one row per record as well as for the header too. It might look
: like
: multiple rows in your text editor but it is technically just one row.
|
D******n
发帖数: 2836 | 13 it is called text/line/word wrap
technically
SAS
【在 a********6 的大作中提到】
: "It might look like multiple rows in your text editor but it is technically
: just one row."
: 是啊，你的意思是CSV文件其实可以有记忆，将看起来多行的东西记为一行？而且，SAS
: 可以识别这种记忆？谢谢。
|
a********6
发帖数: 14468 | 14 长见识了，谢谢。请问大哥，学SAS起步阶段，是看网上video好，还是SAS提供的help
好，还是bbs上的，或者大学的讲义？
【在 D******n 的大作中提到】
: it is called text/line/word wrap
:
: technically
: SAS
|
