p********a 发帖数: 5352 | 1 ☆─────────────────────────────────────☆
madrid (马甲) 于 (Fri May 30 14:07:01 2008) 提到:
My model is a constrained regression ( a linear regression model with
inequality constraints)...Thank you.
☆─────────────────────────────────────☆
tosi (我的名字叫/tu'zi:/) 于 (Fri May 30 14:10:22 2008) 提到:
Sure you can.
(1) Specify the two models correctly and fit seperately.
(2) Retrieve the log likelihood form FitStatistics ODS table and do the math
by yourself.
SAS does not compute LRT for you.
☆────── |
|
j*****7 发帖数: 4348 | 2 请问如何suppress在图中的logrank test的结果? |
|
s**c 发帖数: 1247 | 3 用ods statement,输出保存到一个data set |
|
x**g 发帖数: 807 | 4 不是很清楚你的问题。如果你要选择特定的Output的部分,可以查看ODS(output
delivery system)的使用. |
|
r********e 发帖数: 100 | 5 I will look up the ODS reference.
Thank you xtgg! |
|
m*****8 发帖数: 654 | 6 I try to use proc report to do the listings, and i have trouble of ODS, page
count etc. I try to find some trick but can't get much.
So i review the other programmer's codes and found most of them use data
steps instead of proc report.
If that proc report is not that popular in pharma TFLs? |
|
c****g 发帖数: 156 | 7 most of our TFL are done by proc report, you need to do something extra when
use with ODS |
|
g**r 发帖数: 425 | 8 Look at SUGI proceedings, there are a few articles talking about it.
With SAS version getting newer, ODS is also getting fancier.
page |
|
p********a 发帖数: 5352 | 9 1. If it is variable label, you can use the split option
2. use width after the define statement
3. I am lazy too, sorry |
|
m*****8 发帖数: 654 | 10 i figure q3 out, break after / page;
but i got new tricky questions, I group by the id, and if the list of that
id longer than a page i need to print on the first line of next page, how
can i get that?
original:
id value
1 1
" " 2
" " ...
" " 12
and 8 is page break, and i need:
id value
1 1
" " 2
" " ...
" " 8
page break
1 9
" " ...
" " 12
thanks big shrims |
|
m*****8 发帖数: 654 | 11 unfortuanately, the programmers in my company do this in data steps and
macros instead of proc report, which is way more complicated... |
|
|
s*********e 发帖数: 1051 | 13 2 "proc contents" with ods output... and 1 "proc compare" |
|
x**g 发帖数: 807 | 14 proc content的结果是用ODS 保存的吧?把他改掉就好了。 |
|
s*********e 发帖数: 1051 | 15 跑两个PROC CONTENTS加ODS OUTPUT以后,一个PROC COMPARE就全搞定了。 |
|
r******n 发帖数: 54 | 16 唉,只有papertigar明白了我的意思。
我说了,别人的pdf文件就是final result了,根本就没有人家的program和dataset。
人家就只送了个pdf文件以及data过来,我要做得就是make sure这个data是与他们run
的proc contents(pdf)结果是一致的。所以完全不可能重新改ods.不然的话,那不太
简单了。 |
|
|
l*******l 发帖数: 204 | 18 try export or ODS. Please post the whole log file so other may help. |
|
y****1 发帖数: 400 | 19 我要文本文档的格式所以只能用ods rtf,但是它写到word里面也是有SAS输出的格式的
也不行。
好烦啊,为啥写不全啊!!!
烦死了烦死了。 |
|
s*r 发帖数: 2757 | 20 array + call execute ?
ods ouput |
|
D******n 发帖数: 2836 | 21 在proc logistic里面画roc好像不是很直接。要用ods html来着。
不过这是SAS一向的怪僻。 |
|
c*********t 发帖数: 340 | 22 macro
里面包括proc reg XXX;
ods output ParameterEstimates=result1
然后merge 之前的result和新的Result1
最后输出成一个文件
我这几天也都在做类似的事情,反正这种思路是可行的~ |
|
|
x***n 发帖数: 103 | 24 另外这个好像proc freq 没问题,proc tabulate就有问题了 |
|
y****1 发帖数: 400 | 25 是哦……
嘻嘻。还蛮有意思的。
感觉发现SAS的bug还蛮牛的。
and
when |
|
p********a 发帖数: 5352 | 26 SUGI上好多TALK都是很基本的。。。比方说10 BASIC SKILLS FOR SAS/SQL。。。
SAS这个东西,定义高手很困难,因为SAS其实方向太多,你是MACRO的高手,可能你XML
、ODS搞不过别人,你是SCL的高手,可能你SQL不行。何况这些方方面面不需要都掌握
的。另外,SAS和行业结合很深,一个行业的高手可能在另一个行业需要好长时间PICK
UP
一般来说,只要把MACRO,BASE,SQL熟练掌握,可以算准高手了,如果再在本行业熟练
运用,那就可以算高手了 |
|
q**j 发帖数: 10612 | 27 if column:
look at tagattr
proc report data = yourdata nowd;
define money / style(column) = {tagattr =$#,##0};
define percentage / style(column) = {tagattr =0%};
define order / style(column)=[background = Light Orange];
run;
if row:
use compute block and do something similar. |
|
s********r 发帖数: 729 | 28 不是column,是我之前问过的自定义cell的format。
很伤脑筋。 |
|
a*****8 发帖数: 110 | 29 I am not familiar with proc report, but I think all those formats are for
the SAS output, not excel. Maybe you can try output the table into a SAS
dataset, then convert the numeric value into character using
char=put (num, COMMAw.d);
then output the character value into excel |
|
q**j 发帖数: 10612 | 30 定义列的格式就是用我说的format. 用 {tagattr ='#,##0'}
其他的格式你用excel做好,存成xml,就可以在特定的地方看到这样的格式写法。
定义行的格式我没有试过,应该类似。你给sas打电话好了。 |
|
q**j 发帖数: 10612 | 31 you can also read this doc.
http://www.excursive.com/sas/ProcReportPaper.pdf
whatever you want to do can be done if you read after page 15 of this doc. y
ou can even format each single cell in the excel report. mostly like you wil
l use compute block plus call define(_row_, whatever). |
|
s********r 发帖数: 729 | 32 Sorry for the late reply. I actually contacted the sas support and get some
answer similar to yours. It still work on some format. I am trying to figure
those that are not proper.
Thanks you!
y
wil |
|
|
m******1 发帖数: 19713 | 34 Thanks.
It helps with outputing likelihood ratio test statistic. What about score
test?
If it is an univariate model, score and LR tests can be outputed using ods using the global test results. However, my model is a multivariable model. |
|
P****D 发帖数: 11146 | 35 Not sure that I follow you... But try ODS TRACE your PROC CONTENTS and
figure out the outputable dataset you need. |
|
s*********e 发帖数: 1051 | 36 ods output position = output;
proc contents data = yourdata varnum;
run; |
|
w***y 发帖数: 114 | 37 ods output parameterEstimate-est;
proc logistic;
by hospital;
model y=x1 x2 x3;
output .........;
run; |
|
s*******2 发帖数: 791 | 38 1.用ODS 或者 Proc Export,只知道怎么创建一个ExcelBook.有什么方法可以创建多个
Tab吗?例如有什么options? 或者file format?
2.我有一组SAS数据要输出到Excel.希望按照变量A的数值(A=1001, A=1002, A=1003,..
.),为每一个可能的A值输出一个sheet在这个Excel book。也就是说如果A有10个值,
这个ExcelBook就含有10 个Tabs。怎么实现?谁能给我一个Macro?
谢谢 |
|
p**********e 发帖数: 92 | 39 Use
ods tagsets.excelxp ...... |
|
|
d*****6 发帖数: 100 | 41 下周有个1小时的电话面试,求建议如何准备?这机会对我很重要,请各位给些建议!
!多谢了!!
还有 什么叫ETL processes??
职位:clinical SAS programmer
单位:Heath care system
面试人: biastatistician
我教育背景:MS in biostatistics
job description: must have strong SAS programming skills with respect to
ETL processes (data step, PROC SQL), report writing (SAS/GRAPH,PROC
REPORT, PROC TABULATE, ODS).Familiarity with basic statistical procedures
is required. |
|
|
f*****a 发帖数: 156 | 43 这段SAS code转成R应该怎么写?
proc mixed data=temp;
by x1 x2 x3;
class x4 x5 x6 x7 x8;
model y_&&t&i = /s;
random x4*x5 x6 x7(x6) x4 x6*x4 x8 /s;
ods output solutionf=AAA solutionr=BBB covparms=cov; run;
run;
试着写了个,好像不行,不一样:
fit1=lmer(y~1+(1|x4:x5)+(1|x6)+(x6/x7)+(1|x4)+(1|x6:x4)+(1|x8),data=temp)
summary(fit1)
fit2=lme(y~1,random=list(~1|x4:x5,~1|x6,~x6/x7,~1|x4,~1|x6:x4,~1|x8),data=
temp)
summary(fit2) |
|
d*******1 发帖数: 854 | 44 SAS有五大体系, BASE, macro, GRAPH, STAT, SQL, 如果都能达倒信手捏来的功力, 在
懂一点ODS的话, 那基本上一台笔记本电脑就可以在数据分析界横行天下了 |
|
|
z****e 发帖数: 2024 | 46 Which of the following choices is an unacceptable ODS
destination for producing output that can be viewed in Microsoft Excel?
A. MSOFFICE2K
B. EXCELXP
C. CSVALL
D. WINXP |
|
q**j 发帖数: 10612 | 47 ole db or ods tagset.xp, both works. |
|
B***y 发帖数: 138 | 48 我需要把data list和freq output 放在Excel 的两个不同sheet 里。
用ods tagsets完全可以做到,我的问题是,我需要对两个sheet 有不同的style.
看上去,excelxp只容许我用同一个style而且必须放在第一个excelxp里。我做了一些
research但是没发现更好的办法。我的第二个sheet里有很多proc freq的 output,
proc freq不象 proc report 或者 print那样有很多variable cell format的options.
那里可以override 在excelxp里设置的style. 我想知道,另个不同style 在excelxp
tag里是不是doable?
谢谢 |
|
|
q********i 发帖数: 795 | 50 second that
ods output ParameterEstimates=qiqiandlei |
|