Creating a new variable by dividing its response by the mean of the original variable

[mblack-732]

Hello!

I'm trying to create a new variable that is a "standardized" response of the original variable... that is every observation had a response value for a variable (for the sake of this question, that variable's name is "old"). Is there a way to create a new variable (called "new") such that

new observation response = (old observation response * mean old response)/standard deviation of old response

I understand that I can find these numbers by running a proc means and inputting them into these formulas... but then I have to round. Is there a way to do it without that?

Thank you!

[ballardw]

Sounds like you want Proc STDIZE. It has many standardization methods.

And example:

proc stdize data=sashelp.class out=work.classstd
     oprefix
     sprefix=new
     method=mean;
     var height;
run;

This standardizes Height using the mean. The Oprefix option keeps the original variable (Height), sprefix prefixes any standardized variable names with the prefix "new", creating newheight. If you have multiple variables on the var statement then you would have the original variable and a "new" standardized version for each.

If neither the Oprefix or Sprefix is used the output standardized variable will have the original variable name. If you omit Oprefix but use Sprefix to name a new variable the original variable is not n the output data. The Oprefix can also add a prefix to the original variable using Oprefix=Old for example.

[novinosrin]

Sir @ballardw  Thank you for the notes. Super and priceless:)

[mblack-732]

Hi! Thank you for your response! Is there a way in proc standardize for it to be multiplied by mean and then divided by the standard deviation (not just standardized by mean)? what is the specific command for that?

Thanks again!

[ballardw · Accepted solution]

---

@mblack-732 wrote:

Hi! Thank you for your response! Is there a way in proc standardize for it to be multiplied by mean and then divided by the standard deviation (not just standardized by mean)? what is the specific command for that?

Thanks again!

---

Please read the documentation for Proc STDIZE. There are lots of options:

METHOD= option.

Table 106.2: Available Standardization Methods

Method	Location	Scale
MEAN	Mean	1
MEDIAN	Median	1
SUM	0	Sum
EUCLEN	0	Euclidean length
USTD	0	Standard deviation about origin
STD	Mean	Standard deviation
RANGE	Minimum	Range
MIDRANGE	Midrange	Range/2
MAXABS	0	Maximum absolute value
IQR	Median	Interquartile range
MAD	Median	Median absolute deviation from median
ABW(c)	Biweight one-step M-estimate	Biweight A-estimate
AHUBER(c)	Huber one-step M-estimate	Huber A-estimate
AWAVE(c)	Wave one-step M-estimate	Wave A-estimate
AGK(p)	Mean	AGK estimate (ACECLUS)
SPACING(p)	Mid-minimum spacing	Minimum spacing
L(p)	L(p)	L(p)
IN(ds)	Read from data set	Read from data set

STD is what you are asking for. This is the default if no METHOD= option is provided.