EViews Help: @cumdp

@cumdp

Cumulative process of positive (above threshold) changes.

Compute the partial sum process of positive (above the threshold y) changes in the series x beginning in the specified date.

Syntax: @cumdp(x, d[, y, s])

x: series

d: string

y: (optional) number

s: (optional) sample string or object

Return: series

Consider the partial sum decomposition of a variable

given a initial value

where

, and

are the partial sum processes of the differences for positive, negative, and zero changes in

relative to the threshold y:

This function returns the positive partial sums

for the current or specified sample.

• The date d string specification determines

• Values for dates prior to d will be NAs.

• The optional

argument specifies the threshold value. By default

This function is panel aware.

Examples

The code below produces a graph of a sine wave x and @cumdp applied on x.

wfcreate u 50

series x = @sin(@pi*@trend/4)

group g x @cumdp(x,1)

g.line

Cross-references

See also @cumdn, @cumdz, @dcumdp, @dcumdn, and @dcumdz.