On Mon, 16 May 2016, Fernando Fernandes Neto wrote:
Hello folks.
I'd like to know if there is a simple way of making a fractional
integration of a time series.
I.e., I've built a model for a fractional differenced series. Then, I'd
like to obtain the original level series again, for comparison purposes
with other traditional models. How can I do that?
The fracdiff() function accepts negative orders. For example
<hansl>
nulldata 10
set seed 666
a = normal()
b = fracdiff(a, 0.5)
c = fracdiff(b, -0.5)
print a b c -o
</hansl>
yields
<output>
a b c
1 0.199469 0.199469 0.199469
2 1.328938 1.229204 1.328938
3 -2.120188 -2.809591 -2.120188
4 -0.846691 0.034819 -0.846691
5 -0.286455 0.311063 -0.286455
6 -0.284065 0.040144 -0.284065
7 0.394352 0.667501 0.394352
8 0.160543 0.077359 0.160543
9 -0.306122 -0.364137 -0.306122
10 -0.057199 0.102028 -0.057199
</output>
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------