Trimmed mean.
Compute the mean after the p-percent largest and smallest values have been removed.
Syntax: @trmean(x, p[, s])
x: series, vector, matrix
p: number, series, vector, matrix
s: (optional) sample string or object when x is a series and assigning to a series
Return: number
For
x with
observations and
, define the number of trimmed end observations as
, where
is the integer floor function.
Then the trimmed mean is given by
where the order statistics
represent data for the
observations ordered from low to high,
For series calculations, EViews will use the current or specified workfile sample.
Examples
= @trmean(y, 5)
computes the 5% trimmed mean for series y.
Consider the commands
matrix m = @mrnd(10,10)
m(1,1) = 1000
scalar om = @mean(m)
scalar trm = @trmean(m,1)
The mean of M should be approximately 10.495, whereas the trimmed mean (with the largest and smallest 1% trimmed off) should be close to 0.5.
Cross-references
