大数据

SAS Miss 用法

2019-01-08  本文已影响10人  不连续小姐

SAS Day 20: Miss function

There are two miss functions in SAS I really like, NMISS and CMISS.

Problem:

if we want to calculate the duration as aendt(Adverse Event End Date) — aestdt (Adverse Event Start Date). Ideally, we can use adurn=aendt-aestdt +1. However, in reality, we know there are high probabilities with missing AE dates.

Background Knowledge:

NMISS function is reserved for numeric variables.

MISSING and CMISS function can be used with either character or numeric arguments.

NMISS and CMISS count the number of arguments with missing values.

[caption id="attachment_1385" align="alignnone" width="500"] image

Engin_Akyurt / Pixabay[/caption]

Solution:

  1. Basic Approach
if aendt^=. and aestdt^=. then adurn=aendt-aestdt+12

2. NMISS function

if nmiss(aendt, astdt)=0 then
adurn=aendt-astdt+1

What if we want the Adverse Event Duration Time of the Patients’ with AE ends up in hospitalization?

3. CMISS function

set ae;
if cmiss(aeshosp, aestdy, aeendt)=0
then adurn=aestdy-aeendt+1;
run;
image

Summary:

Methods of selecting non-missing records:

1. x neq. , x neq “ “, Miss(x)=0
note: with Miss function, we don’t need to worry about the type of variables

2. NMISS: good for selecting multiple numbers of non-missing numerical variables.
e.g. nmiss(x1, x2,….xn)=0

3. CMISS: good for selecting multiple numbers of non-missing character or mixed variables.
e.g. cmiss(subjid, aestdy, aeendt)=0

Happy Studying!!! ❄

上一篇下一篇

猜你喜欢

热点阅读