ARFIMA (long memory) models

Christopher F Baum

EC 327: Financial Econometrics

Boston College, Spring 2013

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

1 / 44

ARFIMA (long memory) models

ARFIMA (long memory) models

In estimating an ARIMA model, the researcher chooses the integer
order of differencing d to ensure that the resulting series (1 L)d yt is a
stationary process.
As unit root tests often lack the power to distinguish between a truly
nonstationary (I(1)) series and a stationary series embodying a
structural break or shift, time series are often first-differenced if they do
not receive a clean bill of health from unit root testing.
Many time series exhibit too much long-range dependence to be
classified as I(0) but are not I(1). The ARFIMA model is designed to
represent these series.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

2 / 44

ARFIMA (long memory) models

This problem is exacerbated by reliance on DickeyFuller style tests,

including the improved ElliottRothenbergStock (Econometrica, 1996,
dfgls) test, which have I(1) as the null hypothesis and I(0) as the
alternative. For that reason, it is a good idea to also employ a test with
the alternative null hypothesis of stationarity (I(0)) such as the
KwiatkowskiPhillipsSchmidtShin (J. Econometrics, 1992, kpss)
test to see if its verdict agrees with that of the DickeyFuller style test.
The KPSS test, with a null hypothesis of I(0), is also useful in the
context of the ARFIMA model we now consider. This model allows for
the series to be fractionally integrated, generalizing the ARIMA models
integer order of integration to allow the d parameter to take on
fractional values, 0.5 < d < 0.5.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

3 / 44

ARFIMA (long memory) models

The concept of fractional integration is often referred to as defining a

time series with long-range dependence, or long memory. Any pure
ARIMA stationary time series can be considered a short memory
series. An AR(p) model has infinite memory, as all past values of t
are embedded in yt , but the effect of past values of the disturbance
process follows a geometric lag, damping off to near-zero values
quickly. A MA(q) model has a memory of exactly q periods, so that the
effect of the moving average component quickly dies off.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

4 / 44

ARFIMA (long memory) models

The ARFIMA model

The ARFIMA model1

The model of an autoregressive fractionally integrated moving average
process of a timeseries of order (p, d, q), denoted by ARFIMA
(p, d, q), with mean , may be written using operator notation as
(L)(1 L)d (yt ) = (L)t , t i.i.d.(0, 2 )
where L is the backward-shift operator, (L) = 1 - 1 L - .. - p Lp , (L)
= 1 + 1 L + ... + q Lq , and (1 L)d is the fractional differencing
operator defined by
(1 L)d =

X
k =0

(k d)Lk
(d)(k + 1)

with () denoting the gamma (generalized factorial) function. The

parameter d is allowed to assume any real value.
1

See Baum and Wiggins (Stata Tech.Bull., 2000).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

5 / 44

ARFIMA (long memory) models

The ARFIMA model

The arbitrary restriction of d to integer values gives rise to the standard

autoregressive integrated moving average (ARIMA) model. The
stochastic process yt is both stationary and invertible if all roots of (L)
and (L) lie outside the unit circle and |d| < 0.5. The process is
nonstationary for d 0.5, as it possesses infinite variance; see
Granger and Joyeux (JTSA, 1980).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

6 / 44

ARFIMA (long memory) models

The ARFIMA model

Assuming that d [0, 0.5), Hosking (Biometrika, 1981) showed that

the autocorrelation function, (), of an ARFIMA process is
proportional to k 2d1 as k . Consequently, the autocorrelations of
the ARFIMA process decay hyperbolically to zero as k in
contrast to the faster, geometric decay of a stationary ARMA process.
Pn

For d (0, 0.5), j=n |(j)| diverges as n , and the ARFIMA

process is said to exhibit long memory, or long-range positive
dependence. The process is said to exhibit intermediate memory
(anti-persistence), or long-range negative dependence, for
d (0.5, 0).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

7 / 44

ARFIMA (long memory) models

The ARFIMA model

The process exhibits short memory for d = 0, corresponding to

stationary and invertible ARMA modeling. For d [0.5, 1) the process
is mean reverting, even though it is not covariance stationary, as there
is no long-run impact of an innovation on future values of the process.
If a series exhibits long memory, it is neither stationary (I(0)) nor is it a
unit root (I(1)) process; it is an I(d) process, with d a real number.
A series exhibiting long memory, or persistence, has an autocorrelation
function that damps hyperbolically, more slowly than the geometric
damping exhibited by short memory (ARMA) processes. Thus, it may
be predictable at long horizons. An excellent survey of long memory
modelswhich originated in hydrology, and have been widely applied
in economics and financeis given by Baillie (J. Econometrics, 1996).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

8 / 44

ARFIMA (long memory) models

Approaches to estimation of the ARFIMA model

Approaches to estimation of the ARFIMA model

There are two approaches to the estimation of an ARFIMA (p, d, q)

model: exact maximum likelihood estimation, as proposed by Sowell
(1992), and semiparametric approaches. Sowells approach requires
specification of the p and q values, and estimation of the full ARFIMA
model conditional on those choices. This involves the challenge of
choosing an appropriate ARMA specification.
We first describe semiparametric methods, in which we assume that
the short memory or ARMA components of the timeseries are
relatively unimportant, so that the long memory parameter d may be
estimated without fully specifying the data generating process.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

9 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The Lo Modified Rescaled Range estimator2

The Stata routine lomodrs performs Los (Econometrica, 1991)
modified rescaled range (R/S, range over standard deviation) test for
long range dependence of a time series. The classical R/S statistic,
devised by Hurst (1951) and Mandelbrot (AESM, 1972), is the range of
the partial sums of deviations of a timeseries from its mean, rescaled
by its standard deviation. For a sample of n values {x1 , x2 , . . . xn },

k
k
X
X




1
Max1k n
Qn =
xj xn Min1k n
xj xn
sn
j=1

j=1

where sn is the maximum likelihood estimator of the standard deviation

of x.
2

See Baum and Rom (Stata Tech. Bull., 2001).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

10 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The first bracketed term is the maximum of the partial sums of the first
k deviations of xj from the full-sample mean, which is nonnegative.
The second bracketed term is the corresponding minimum, which is
nonpositive. The difference of these two quantities is thus nonnegative,
so that Qn > 0. Empirical studies have demonstrated that the R/S
statistic has the ability to detect long-range dependence in the data.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

11 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Like many other estimators of long-range dependence, though, the

R/S statistic has been shown to be excessively sensitive to
short-range dependence, or short memory, features of the data. Lo
(1991) shows that a sizable AR(1) component in the data generating
process will seriously bias the R/S statistic. He modifies the R/S
statistic to account for the effect of short-range dependence by
applying a NeweyWest correction (using a Bartlett window) to derive
a consistent estimate of the long-range variance of the timeseries.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

12 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

For maxlag> 0, the denominator of the statistic is computed as the

NeweyWest estimate of the long run variance of the series. If
maxlag is set to zero, the test performed is the classical
HurstMandelbrot rescaled-range statistic. Critical values for the test
are taken from Lo, 1991, Table II.
Inference from the modified R/S test for long range dependence is
complementary to that derived from that of other tests for long
memory, or fractional integration in a timeseries, such as kpss,
gphudak, modlpr and roblpr.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

13 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The GewekePorter-Hudak log periodogram

regression (LPR) estimator

The Stata command gphudak (Baum and Wiggins, Stata Tech. Bull.,
2000) performs the Geweke and Porter-Hudak (JTSA, 1983)
semiparametric log periodogram regression, often described as the
GPH test, for long memory (fractional integration) in a timeseries.
The GPH method uses nonparametric methodsa spectral regression
estimatorto evaluate d without explicit specification of the short
memory (ARMA) parameters of the series. The series is usually
differenced so that the resulting d estimate will fall in the [-0.5, 0.5]
interval.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

14 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Geweke and Porter-Hudak (1983) proposed a semiparametric

procedure to obtain an estimate of the memory parameter d of a
fractionally integrated process Xt in a model of the form
(1 L)d Xt = t ,
where t is stationary with zero mean and continuous spectral density
f () > 0.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

15 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The estimate d is obtained from the application of ordinary least

squares to

2


log (Ix (s )) = c d log 1 eis  + residual
computed
over the fundamental
frequencies


1 Pn
2s
its

X
e
s = n , s = 1, ..., m, m < n . We define x (s ) =
t
t=1
2n
as the discrete Fourier transform (dft) of the timeseries Xt , 


i
s

Ix (s ) = x (s ) x (s ) as the periodogram, and xs = log 1 e  .
Ordinary least squares on (7) yields
Pm
s=1 xs log Ix (s )

d = 0.5
.
Pm
2
s=1 xs

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

16 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Various authors have proposed methods for the choice of m, the

number of Fourier frequencies included in the regression. The
regression slope estimate is an estimate of the slope of the series
power spectrum in the vicinity of the zero frequency; if too few
ordinates are included, the slope is calculated from a small sample. If
too many are included, medium and high-frequency components
of the

spectrum will contaminate the estimate. A choice of T , or power =

0.5 is often employed.
To evaluate the robustness of the GPH estimate, a range of power
values (from 0.400.75) is commonly calculated as well. Two
estimates of the d coefficients standard error are commonly
employed: the regression standard error, giving rise to a standard
t-test, and an asymptotic standard error, based upon the theoretical
2
variance of the log periodogram of 6 . The statistic based upon that
standard error has a standard normal distribution under the null.
Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

17 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The Phillips Modified GPH log periodogram

regression estimator

The Stata routine modlpr (Baum and Wiggins, Stata Tech. Bull.,
2000) computes a modified form of the GPH estimate of the long
memory parameter, d, of a timeseries, proposed by Phillips (Cowles,
1999a, 1999b). Phillips (1999a) points out that the prior literature on
this semiparametric approach does not address the case of d = 1, or a
unit root, in (6), despite the broad interest in determining whether a
series exhibits unit-root behavior or long memory behavior, and his
work showing that the d estimate of (7) is inconsistent when d > 1,
with d exhibiting asymptotic bias toward unity.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

18 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

This weakness of the GPH estimator is solved by Phillips Modified Log

Periodogram Regression estimator, in which the dependent variable is
modified to reflect the distribution of d under the null hypothesis that
d = 1. The estimator gives rise to a test statistic for d = 1 which is a
standard normal variate under the null.
Phillips suggests that deterministic trends should be removed from the
series before application of the estimator. Accordingly, the routine will
automatically remove a linear trend from the series. This may be
suppressed with the notrend option. The comments above regarding
power apply equally to modlpr.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

19 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Phillips (1999b) modification of the GPH estimator is based on an

exact representation of the dft in the unit root case. The modification
expresses
u (s )
eis
Xn

x (s ) =

i
i
s
s
1e
1e
2n
and the modified dft as
Xn
eis

x (s ) = x (s ) +
1 eis 2n
with associated periodogram ordinates Iv (s ) = x (s ) x (s )
(1999b, p.9). He notes that both x (s ) and, thus, Iv (s ) are
observable functions of the data.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

20 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The log-periodogram
regression is now
the regression of log Iv (s ) on


P
= m1 m

as = log 1 eis  . Defining a

s=1 as and xs = as a, the
modified estimate of the long-memory parameter becomes
Pm
s=1 xs log I (s )

.
d = 0.5
Pm
2
s=1 xs

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

21 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Phillips proves that, with appropriate assumptions on the distribution of

t , the distribution of d follows




2

m d d d N 0,
,
24
so that d has the same limiting distribution at d = 1 as does the GPH
estimator in the stationary case so that d is consistent for values of d
around unity. A semiparametric test statistic for a unit root against a
fractional alternative is then based upon the statistic (1999a, p.10):


m d 1

zd =
/ 24
with critical values from the standard normal distribution. This test is
consistent against both d < 1 and d > 1 fractional alternatives.
Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

22 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Robinsons Log Periodogram Regression

estimator

The Stata routine roblpr (Baum and Wiggins, Stata Tech. Bull.,
2000) computes the Robinson (Ann. Stat.,1995) multivariate
semiparametric estimate of the long memory (fractional integration)
parameters, d(g), of a set of G timeseries, y (g), g = 1, G with G 1.
When applied to a set of timeseries, the d(g) parameter for each
series is estimated from a single log-periodogram regression which
allows the intercept and slope to differ for each series.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

23 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

One of the innovations of Robinsons estimator is that it is not

restricted to using a small fraction of the ordinates of the empirical
periodogram of the series: that is, the reasonable values of power
need not exclude a sizable fraction of the original sample size. The
estimator also allows for the removal of one or more initial ordinates,
and for the averaging of the periodogram over adjacent frequencies.
The rationales for using non-default values of either of these options
are presented in Robinson (1995).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

24 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Robinson (1995) proposes an alternative log-periodogram regression

estimator which he claims provides modestly superior asymptotic
efficiency to d (0) (d (0) being the Geweke and Porter-Hudak
estimator) (1995, p.1052).
Importantly, Robinsons formulation of the log-periodogram regression
also allows for the formulation of a multivariate model, providing
justification for tests that different time series share a common
differencing parameter. Normality of the underlying time series is
assumed, but Robinson claims that other conditions underlying his
derivation are milder than those conjectured by GPH.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

25 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

We present here Robinsons multivariate formulation, which applies to

a single time series as well. Let Xt represent a Gdimensional vector
with g th element
, g = 1, ..., G. Assume that Xt has a spectral
R Xgt ij
density matrix e f () d, with (g, h) element denoted as fgh () .
The g th diagonal element, fgg () , is the power spectral density of Xgt .
For 0 < Cg < and 21 < dg < 21 , assume that fgg () Cg 2dg as
0+ for g = 1, ..., G. The periodogram of Xgt is then denoted as

2
n
X


1
it 
Ig () = (2n) 
Xgt e  , g = 1, ...G


t=1

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

26 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Without averaging the periodogram over adjacent frequencies nor

omission of l initial frequencies from the regression, we may define
Ygk = log Ig (k ) . The least squares estimates of c = (c1 , ...cG )0 and
d = (d1 , ..., dG )0 are given by


o
n


c
1
0
0
,
=
vec
Y
Z
Z
Z

d
where Z = (Z1 , ...Z
m )0 , Zk = (1, 2 log k )0 , Y = (Y1 , ...YG ) , and
0
Yg = Yg,1 , ...Yg,m for m periodogram ordinates.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

27 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Standard errors for dg and for a test of the restriction that two or more
of the dg are equal may be derived from the estimated covariance
matrix of the least squares coefficients. The standard errors for the
estimated parameters are derived from a pooled estimate of the
variance in the multivariate case, so that their interval estimates differ
from those of their univariate counterparts. Modifications to this
derivation when the frequency-averaging (j) or omission of initial
frequencies (l) options are selected may be found in Robinson (1995).

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

28 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

Maximum likelihood estimators of ARFIMA models

The official Stata command arfima implements the full maximum

likelihood estimation of the ARFIMA(p,d,q) model, as proposed by
Sowell (J. Econometrics, 1992). The ARFIMA model has the d
parameter to handle long-run dependence and ARMA parameters to
handle short-run dependence. Sowell has argued that using different
parameters for different types of dependence facilitates estimation and
interpretation.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

29 / 44

ARFIMA (long memory) models

Semiparametric estimators for I(d) series

The ARFIMA model specifies that

yt = (1 L)d ((L))1 (L)t
After estimation, the short-run effects are obtained by setting d = 0,
and describe the behavior of the fractionally differenced process
(1 L)d yt . The long-run effects use the estimated value of d, and
describe the behavior of the fractionally integrated yt .
Granger and Joyeux (1980) motivate ARFIMA models by noting that
their implied spectral densities for d > 0 are finite except at frequency
0, whereas stationary ARMA models have finite spectral densities at
all frequencies. The ARFIMA model is able to capture the long-range
dependence, which cannot be expressed by stationary ARMA models.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

30 / 44

ARFIMA (long memory) models

Applications

lomodrs and classical rescaled range estimators

Data from Terence Mills Econometric Analysis of Financial Time
Series on returns from the annual S&P 500 index of stock prices,
1871-1997, are analyzed.

. use http://fmwww.bc.edu/ec-p/data/Mills2d/SP500A.DTA, clear

. lomodrs sp500ar
Lo Modified R/S test for sp500ar
Critical values for H0: sp500ar is not long-range dependent
90%: [ 0.861, 1.747 ]
95%: [ 0.809, 1.862 ]
99%: [ 0.721, 2.098 ]
Test statistic:

.781

Christopher F Baum (BC / DIW)

(1 lags via Andrews criterion)

ARFIMA (long memory) models

N = 124

Boston College, Spring 2013

31 / 44

ARFIMA (long memory) models

Applications

. lomodrs sp500ar, max(0)

Hurst-Mandelbrot Classical R/S test for sp500ar
Critical values for H0: sp500ar is not long-range dependent
90%: [ 0.861, 1.747 ]
95%: [ 0.809, 1.862 ]
99%: [ 0.721, 2.098 ]
Test statistic:
.799
N = 124
. lomodrs sp500ar if tin(1946,)
Lo Modified R/S test for sp500ar
Critical values for H0: sp500ar is not long-range dependent
90%: [ 0.861, 1.747 ]
95%: [ 0.809, 1.862 ]
99%: [ 0.721, 2.098 ]
Test statistic:

1.29

Christopher F Baum (BC / DIW)

(0 lags via Andrews criterion)

ARFIMA (long memory) models

N = 50

Boston College, Spring 2013

32 / 44

ARFIMA (long memory) models

Applications

For the full sample, the null of stationarity may be rejected at 95%
using either the Lo modified R/S statistic or the classic
HurstMandelbrot statistic. For the postwar data, the null may not be
rejected at any level of significance. Long-range dependence, if
present in this series, seems to be contributed by pre-World War II
behavior of the stock price series.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

33 / 44

ARFIMA (long memory) models

Applications

GPH, Phillips modlpr, Robinson roblpr

Data from Terence Mills Econometric Analysis of Financial Time
Series on UK FTA All Share stock returns (ftaret) and dividends (ftadiv)
are analyzed.

. use http://fmwww.bc.edu/ec-p/data/Mills2d/FTA.DTA, clear

. gphudak ftaret,power(0.5 0.6 0.7)
GPH estimate of fractional differencing parameter
-----------------------------------------------------------------------------Asy.
Power
Ords
Est d
StdErr t(H0: d=0) P>|t|
StdErr z(H0: d=0) P>|z|
-----------------------------------------------------------------------------.5
20 -.00204
.1603
-0.0127
0.990
.1875
-0.0109
0.991
.6
35 .228244
.1459
1.5645
0.128
.1302
1.7529
0.080
.7
64 .141861
.08992
1.5776
0.120
.09127
1.5544
0.120
------------------------------------------------------------------------------

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

34 / 44

ARFIMA (long memory) models

Applications

. modlpr ftaret, power(0.5 0.55:0.8)

Modified LPR estimate of fractional differencing parameter for ftaret
-----------------------------------------------------------------------------Power
Ords
Est d
Std Err t(H0: d=0)
P>|t|
z(H0: d=1)
P>|z|
-----------------------------------------------------------------------------.5
19
.0231191
.139872
0.1653
0.870
-6.6401
0.000
.55
25
.2519889
.1629533
1.5464
0.135
-5.8322
0.000
.6
34
.2450011
.1359888
1.8016
0.080
-6.8650
0.000
.65
46
.1024504
.1071614
0.9560
0.344
-9.4928
0.000
.7
63
.1601207
.0854082
1.8748
0.065
-10.3954
0.000
.75
84
.1749659
.08113
2.1566
0.034
-11.7915
0.000
.8
113
.0969439
.0676039
1.4340
0.154
-14.9696
0.000
------------------------------------------------------------------------------

. roblpr ftaret
Robinson estimates of fractional differencing parameter for ftaret
------------------------------------------------------Power
Ords
Est d
Std Err t(H0: d=0)
P>|t|
------------------------------------------------------.9
205
.1253645
.0446745
2.8062
0.005
-------------------------------------------------------

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

35 / 44

ARFIMA (long memory) models

Applications

. roblpr ftap ftadiv

Robinson estimates of fractional differencing parameters
Power =
.9
Ords
=
205
------------------------------------------------------------Variable
|
Est d
Std Err
t
P>|t|
-----------------+------------------------------------------ftap
| .8698092
.0163302 53.2640
0.000
ftadiv
| .8717427
.0163302 53.3824
0.000
------------------------------------------------------------Test for equality of d coefficients:
F(1,406) = .00701
Prob > F = 0.9333

. constraint define 1 ftap=ftadiv

. roblpr ftap ftadiv ftaret, c(1)
Robinson estimates of fractional differencing parameters
Power =
.9
Ords
=
205
------------------------------------------------------------Variable
|
Est d
Std Err
t
P>|t|
-----------------+------------------------------------------ftap
| .8707759
.0205143 42.4473
0.000
ftadiv
| .8707759
.0205143 42.4473
0.000
ftaret
| .1253645
.0290116
4.3212
0.000
------------------------------------------------------------Test for equality of d coefficients:
F(1,610) = 440.11
Prob > F = 0.0000
Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

36 / 44

ARFIMA (long memory) models

Applications

The GPH test, applied to the stock returns series, generates estimates
of the long memory parameter that cannot reject the null at the ten
percent level using the t-test.
Phillips modified LPR, applied to this series, finds that d = 1 can be
rejected for all powers tested, while d = 0 (stationarity) may be
rejected at the ten percent level for powers 0.6, 0.7, and 0.75.
Robinsons estimate for the returns series alone is quite precise.
Robinsons multivariate test, applied to the price and dividends series,
finds that each series has d > 0. The test that they share the same d
cannot be rejected. Accordingly, the test is applied to all three series
subject to the constraint that price and dividends series have a
common d, yielding a more precise estimate of the difference in d
parameters between those series and the stock returns series.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

37 / 44

ARFIMA (long memory) models

Applications

Sowell MLE ARFIMA

We model the log of the monthly level of CO above Mauna Loa,

Hawaii, removing seasonal effects by using the twelfth seasonal
difference (S12. in Stata parlance) of that series. We first consider an
ARMA model with a first lag in the AR polynomial and the second lag
in the MA polynomial.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

38 / 44

ARFIMA (long memory) models

Applications

. webuse mloa
. arima S12.log, ar(1) ma(2) vsquish nolog
ARIMA regression
Sample: 1960m1 - 1990m12
Log likelihood =

Number of obs
Wald chi2(2)
Prob > chi2

2001.564
OPG
Std. Err.

S12.log

Coef.

_cons

.0036754

.0002475

.7354346

=
=
=

372
500.41
0.0000

P>|z|

[95% Conf. Interval]

14.85

0.000

.0031903

.0041605

.0357715

20.56

0.000

.6653237

.8055456

.1353086

.0513156

2.64

0.008

.0347319

.2358853

.0011129

.0000401

27.77

0.000

.0010344

.0011914

log

ARMA
ar
L1.
ma
L2.
/sigma

Note: The test of the variance against zero is one sided, and the two-sided
confidence interval is truncated at zero.
. psdensity d_arma omega1

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

39 / 44

ARFIMA (long memory) models

Applications

All parameters are statistically significant, and indicate a high degree

of dependence. This model is nested within the ARFIMA model:
. arfima S12.log, ar(1) ma(2) vsquish nolog
ARFIMA regression
Sample: 1960m1 - 1990m12
Log likelihood =

S12.log

Number of obs
Wald chi2(3)
Prob > chi2

2006.0805

=
=
=

372
248.87
0.0000

OIM
Std. Err.

P>|z|

[95% Conf. Interval]

.003616

.0012968

2.79

0.005

.0010743

.0061578

.2160894

.1015596

2.13

0.033

.0170362

.4151426

.1633916
.4042573

.051691
.080546

3.16
5.02

0.002
0.000

.062079
.2463899

.2647041
.5621246

1.20e-06

8.84e-08

13.63

0.000

1.03e-06

1.38e-06

Coef.

S12.log
_cons
ARFIMA
ar
L1.
ma
L2.
d
/sigma2

Note: The test of the variance against zero is one sided, and the two-sided
confidence interval is truncated at zero.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

40 / 44

ARFIMA (long memory) models

Applications

Here, too, all parameters are significant at the five percent level. The
estimate of d, 0.404, is far from zero, indicating the presence of
long-range dependence. We can compare the models ability to
capture the dynamics of the series by computing their implied spectral
densities over (0, ).
For a stationary time series, the spectral density describes the relative
importance of components at different frequencies. The integral of the
spectral density over (, ) is the variance of the time series.
We can also compute the implied spectral density of the ARFIMA
model, setting d to zero to compute the short-run estimates. The
long-run estimates have infinite density at frequency zero.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

41 / 44

ARFIMA (long memory) models

Applications

All parameters are statistically significant, and indicate a high degree

of dependence. This model is nested within the ARFIMA model:
. psdensity d_arfima omega2
. psdensity ds_arfima omega3, smemory
. line d_arma d_arfima omega1, name(lmem) scheme(s2mono) nodraw ylab(,angle(0))
. line d_arma ds_arfima omega1, name(smem) scheme(s2mono) nodraw ylab(,angle(0)
> )
. graph combine lmem smem, cols(1) xcommon ///
> ti("ARMA and ARFIMA implied spectral densities")
. gr export 82308b.pdf, replace
(file /Users/cfbaum/Dropbox/baum/EC823 S2013/82308b.pdf written in PDF format)

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

42 / 44

ARFIMA (long memory) models

Applications

ARMA and ARFIMA implied spectral densities

5
4
3
2
1
0
0

Frequency
ARMA spectral density

ARFIMA long-memory spectral density

1.5
1
.5
0
0

Frequency
ARMA spectral density

Christopher F Baum (BC / DIW)

ARFIMA short-memory spectral density

ARFIMA (long memory) models

Boston College, Spring 2013

43 / 44

ARFIMA (long memory) models

Applications

The two models imply different spectral densities for frequencies close
to zero when d > 0. The spectral density of the ARMA model remains
finite, but is pulled upward by the models attempt to capture
long-range dependence. The short-run ARFIMA parameters can
capture both low-frequency and high-frequency components in the
spectral density.
In contrast, the ARMA model confounds the long-run and short-run
effects. The added flexibility of the ARFIMA model, with a separate
parameter to capture long-run dependence, is evident in these
estimates.
Although we have not illustrated it here, arfima may also fit
ARFIMA-X models with additional exogenous regressors.

Christopher F Baum (BC / DIW)

ARFIMA (long memory) models

Boston College, Spring 2013

44 / 44