o******6
发帖数: 538 | 1 Some variables in the original excel file are numeric but have no value for
the first 8 observation. When I import the file , these
variables become character variables and their value are all missing. I know
SAS uses the first 8 observations to determine the variable type, does
anyone know how to fix it? Do I need change some SAS setting? Thanks, the only biostatistician in my group didn't help me (no problem in his computer) and I usually figure it out by myself :(.
PROC IMPORT OUT= WORK.a
|
t**i
发帖数: 688 | 2 Can you move the ninth row to the top? |
o******6
发帖数: 538 | 3 It's not a good idea. The same sas code run successfully in my colleague's computer.
He may know how to fix it but he never teach me. I believe it is related to the setting, anyone can help me?
【在 t**i 的大作中提到】
: Can you move the ninth row to the top?
|
f********t
发帖数: 117 | |
o******6
发帖数: 538 | 5 Yes,it works and the variables become character variables. I can use input
function to change them to the numeric variables. But why does the original
code work in my coworker's computer? Is there any simple way to fix it such
as changing the setting?
【在 f********t 的大作中提到】
: maybe try MIXED=YES;
|
p********a
发帖数: 5352 | 6 guessingrows=100; works only for certain higher SAS versions or certain
lower version with parameters changed. |
o******6
发帖数: 538 | 7 Thanks a lot. It doesn't work for my SAS version. My Korea coworker's SAS
version (9.1) is the same as mine. I don't know how he fixes it:( .I may
have to use the stupid way to fix the problem if I cannot figure it out and
I don't like this idea.
【在 p********a 的大作中提到】
: guessingrows=100; works only for certain higher SAS versions or certain
: lower version with parameters changed.
|
o******6
发帖数: 538 | 8 I found the following information. Hope it works.
The solution to this problem is to scan more than the first eight rows. We
don't have a SAS solution; this involves the Windows registry. If you are
comfortable working with the Windows registry, here's the process.
1. Back-up the registry!
2. Find HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Jet\4.0\Engines\Excel.
3. Double click on TypeGuessRows.
4. Change the value to 0. (All rows are scanned when the value is 0.)
for
know
only biostatistician in my
【在 o******6 的大作中提到】
: Some variables in the original excel file are numeric but have no value for
: the first 8 observation. When I import the file , these
: variables become character variables and their value are all missing. I know
: SAS uses the first 8 observations to determine the variable type, does
: anyone know how to fix it? Do I need change some SAS setting? Thanks, the only biostatistician in my group didn't help me (no problem in his computer) and I usually figure it out by myself :(.
: PROC IMPORT OUT= WORK.a
:
|
o******6
发帖数: 538 | 9 我今天试啦，WORK了，嘻嘻，真郁闷我的韩国同事什么都不告诉我，而我还什么都告诉
他，是不是工作了几年的人都这样呀？
【在 o******6 的大作中提到】
: I found the following information. Hope it works.
: The solution to this problem is to scan more than the first eight rows. We
: don't have a SAS solution; this involves the Windows registry. If you are
: comfortable working with the Windows registry, here's the process.
: 1. Back-up the registry!
: 2. Find HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Jet\4.0\Engines\Excel.
: 3. Double click on TypeGuessRows.
: 4. Change the value to 0. (All rows are scanned when the value is 0.)
:
: for
|
