Standardized data (using sample standard deviation).
Return copy of data scaled and translated to have a mean of zero and a Pearson product moment sample standard deviation of one.
Syntax: @stdize(x, [s])
x: series, vector, matrix
s: (optional) sample string or object when x is a series and assigning to a series
Return: series, vector, matrix object
The adjusted data are calculated as
where

and

are the mean and sample standard deviation, respectively, of

.
For series calculations, EViews will use the current or specified workfile sample.
Examples
show @stdize(x)
returns a linked series of the standardized values of x.
Consider the commands
matrix m = @mrnd(100,100)
@mean(m)
@varp(m)
matrix stdm = @stdizep(m)
@mean(stdm)
@varp(stdm)
The matrix M should have mean approximately 0.5 and population variance approximately 0.083. The standardized matrix, STDM, should have mean approximately 0 and population variance of 1.
Cross-references