Scribd
Upload
84 views14 pages

Integrated Time Series and Cointegration: Course: Applied Econometrics Lecturer: Zhigang Li

This document discusses integrated time series and cointegration. It defines integrated processes and cointegration, and explains that two series are cointegrated if they both follow integrated processes of the same order and a linear combination of the two series is integrated of a lower order. The document also discusses unit root processes, spurious regressions, cointegration and error correction models, procedures for testing cointegration like the Engle-Granger method, and limitations of cointegration analysis.

Uploaded by

nekougolo3064
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
84 views14 pages

Integrated Time Series and Cointegration: Course: Applied Econometrics Lecturer: Zhigang Li

This document discusses integrated time series and cointegration. It defines integrated processes and cointegration, and explains that two series are cointegrated if they both follow integrated processes of the same order and a linear combination of the two series is integrated of a lower order. The document also discusses unit root processes, spurious regressions, cointegration and error correction models, procedures for testing cointegration like the Engle-Granger method, and limitations of cointegration analysis.

Uploaded by

nekougolo3064
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 14

Integrated Time Series and

Cointegration

Course: Applied Econometrics


Lecturer: Zhigang Li
Integrated Series and Cointegration
 A series xt is integrated of order d
(we call it I(d) process) if the series
becomes stationary after differencing
d times.
 Two series x and y are cointegrated if
 Both series are of the same order d
 A linear combination of the two series is
integrated to the order b (b<d).
Unit Root Process
 Unit Root Process
 yt=yt-1+ut (ut is a weakly dependent process)
 A random walk (ut is i.i.d. with mean zero) is a
special case of the unit root process.
 No matter how far in the future we look and
how much information we have for the past,
our best prediction of future is today’s value.
 The expected value of a random walk does not
depend on t
 The variance of a random walk increases as a
linear function of time (nonstationary).
 High persistency: Corr(yt, yt+h)=[t/(t+h)]1/2
Spurious Regression
 Spurious regression: X and Y are not related at all but
regressing Y on X shows a significant statistical
correlation between them. This could happen for:
 Omitted variable Z that drives both X and Y
 Trending X and Y
 Even if series Xt and Yt are not trending, the
regression between them may be spurious if X and Y
are independent and are both I(1).
 In this case, the error term of the regression is an I(1)
process, thus strongly dependent, violating consistency
assumptions.
 Solutions
 Include omitted variables
 First difference
 Cointegration
Cointegration and Error Correction
Model
 If β exists such that yt-βxt is an I(0) process, then y and
x are cointegrated.
 yt=α+βxt+e
 A cointegration model between X and Y can be equally
rewritten as
 Δyt=α+γΔxt+δ(yt-1-βxt-1)+u
 While the cointegration model emphasizes the
long-run equilibrium relationship between y and
x, the error correction model characterizes the
short-run adjustment processes towards the
equilibrium relationship.
The Engle-Granger Procedure
 If the series X and Y are integrated to the
same order d, cointegration between X and
Y can be tested through the following two-
stage procedure
 Cointegrating Regression: Regress Y on X (and
other control variables) by OLS
 The residuals from the regression are tested for
the order of integration. If the residuals are
integrated to lower order, then X and Y are
cointegrated.
Rigorous Unit Root Test I
(Dickey-Fuller Test)
 To test whether ρ=1 in yt=α+ρyt-1+e,
rewrite it as Δyt=α+(ρ-1)yt-1+et
 H0: ρ-1=0 H1: ρ-1<0
 The model is assumed to be dynamically
complete (martingale: E(et|yt-1, yt-2,…,y0)=0)
 Because the series yt is I(1) under H0, usual t-
test critical values need to be adjusted (following
Dickey-Fuller) as follows

Significance Level 1% 5% 10%


Critical Value -3.43 -2.86 -2.57
Rigorous Unit Root Test II
(Augmented Dickey-Fuller Test)
 Δyt=α+(ρ-1)yt-1+ Δyt-1+Δyt-2+…+Δyt-p+et
 H0: ρ-1=0 H1: ρ-1<0
 Enough lagged dependent variables are added
so that the model is dynamically complete.
 The lag length is often dictated by the
frequency of the data. For annual data, one or
two lags usually suffice. For monthly data,
twelve lags might be needed.
 The t statistics on the lagged changes have
approximate t distributions, so standard tests
might be used to determine lag lengths.
 The critical values are the same as the Dickey-
Fuller test in last slide.
Rigorous Unit Root Test III
(Dickey-Fuller Test with Time Trend)
 Δyt=α+δt+(ρ-1)yt-1+et
 H0: ρ-1=0 H1: ρ-1<0
 A trend-stationary process can be mistaken for a
unit root process if the time trend is not
controlled for.
 Critical values of the Dickey-Fuller test chanes
when a time trend is included:
Significance Level 1% 5% 10%
Critical Value -3.43 -2.86 -2.57
Critical Value (Trend) -3.96 -3.41 -3.12
Limitations of Cointegration
Analysis
 Pre-test procedures (unit root test of
individual variables) are often
inconclusive.
 There may be substantial small-
sample bias.
 Structural breaks in the time series
can cause difficulties in unit root test
and cointegration analysis.
Aggregated Consumption and the
Demand for Imports (Clarida, 1992)
 Empirical model

m: Import of nonduarable goods


h’: Domestic nondurable goods consumption
p: Relative import prices
v: Stationary disturbance
Pre-test (D-F test) for
Nonstationarity

You might also like