Time Series Econometrics:

Some Basic Concepts

Reference : Gujarati, Chapters 21

• One of the important and frequent types of data used in empirical

analysis.
• But it poses several challenges to econometricians practidoners.
E.g.
 l. Empirical work based on time series data assumes that the
underlying time series is stationary.
2. Autocorreladon: because the underlying time series data is
non-stationary.
3. Spurious/nonsense regression: a very high R2 and significant
regression coefficients (though there is no meaningful
relationship between the two variables)

l . Stochastic Processes
Stationarity Processes ii.
Purely Random Processes lii. Non-
stationary Processes 2. Random
Walk Models
 i. Random Walk with Drift ii.
Random Walk without Drift
3. Unit Root Stochostic Processes
4. Deterministic and Stochastic Trends
5. The Phenomenon of Spurious Regression
6. Tests of Stationarity/non-stationarity
i. Graphical Method ii.
Unit Root Tests

l) Stochastic Processes
• Stochasüc (Random) Process: collection of
random variables ordered in time.
 0 NOTATIONS: Let Y a random vaHable, Y(t) if continuous
(e.g. electrocardiogram), and Yt if discrete (e.g. GDP, PDI,
etc.).
• Now, If we let Y represent GDP, then we can have Yl, Y2, Y3, ,
Y20 where the subscript 1 denotes the 1st observation (i.e. GDP
for the 1st quarter of 1st year) and the subscript 20 denotes the
last observation (i.e. GDP for the 4th quarter of 5th year).
(l) Stochastic Processes

• Stationary Stochastic Processes: A stochastic process is said to be

stationary weakly covariance 2nd-order stationary if:
• Its mean and variance are constant over time, and o The value of the
covariance between the two time periods depends only on the distance/lag
between the two time periods and not the actual time at which the covariance
is computed.

• E.g. let's Yt be a stochastic process, then;

Mean: (1) — Variance:(2)
— Covariance: var (3)

• Where Yk, the covariance (or

auto-covariance) at lag k,

• If k — (), we obtain Yo, which is Simply the variance of Y 02); if k - 1, Y I is

the covariance between two adjacent values of Y

Why are Stationary Time Series so Important?

• Because if a time series is non-staüonary, we can study its
behavior only for the time period under consideration, and as a
consequence, it is not possible to generalize it to other time
periods.
• Therefore, for the purpose of forecasting, such (non-stationary)
time series may be of little practical value.
• Non-stadonary Stochastic Processes: Although our Interest is in
stationary time series, one often encounters nonstationary time
series
A non-stationary time series will have a time-vanång mean or a time-
varymg variance or both.
• We call a stochastic process(time senes) purely random/white noise process if
it has zero mean, constant variance 02, and is serially uncorrelated i.e. [ut
IIDN(O, 02)].

• Note: Here onward, in all equations the assumption of "white noise" will be
applicable on ut .
 2) Random Walk Model (RWM)
• The classic example of non-stationary time senes is
the Random Walk Model (RWM).
• It is often said that asset prices, such as stock prices
or exchange rates, follow a random walk (i.e.
nonstationary).
• Types of Random Walks:
a) Random Walk Without Drift:
I.e. no constant/intercept term and
b) Random Walk With Drift
I.e. a constant term is present
 a) Random Walk without Drift
• The time series Ytis said to be a random walk drift, if
t (4)
• Here, the value of Y at time (t) is equal to its value at time (t
plus a random shock; thus it is an AR(I) model.
• Believers in the Efficient Capital Market Hypothesis argue that
stock prices are essentially random and therefore there is no
scope for profitable speculation in the stock market:
• If one could predict tomorrow's price on the basis of today's
price, we would all be millionaires.
• Now from Yt = Yt_l + ut....... (4) we can write:

Y = Y2 + 113 = Yo + 111 + 112 + 113 and so on...

• In general, if the process started at some time 0 with a value of Yo, we have:

• Therefore,
E(Yt) = but) -Yo (why?)
Because, utis "white noise
• In like fashion, it can be shown that: var (Yt ) = t02 ...... (7)
• Thus, the mean of Y is equal to its starting value, which is constant, but as t
increases, its variance increases indefinitely (thus violating the condition of
stationarity).

• In short, the RWM without drift is a non-stationary stochastic process.

Further Explanation (cont.)

• Now, if you MTite Yt=Yt 1 + lit..... (4) as

(Yt - Yt_l) ..... .. (8)

• It shows that, while Yt is non-stationary, its 1st difference
is stationary.
• In other words, the 1st differences of a random walk time
series are stationary.
 b) Random Walk with Drift
• Let's modify, Yt Yt _ I + ut................(4) as follows:
Yt-l + ... . ..... (9) where
is the drift parameter.
The name drift comes from the fact that if we write the
preceding equation as:
Yt-Y (10)
• It shows that Yt drifts upward/dovvnward, depending on
being positive/negative.
• Note that model Yt = 6 + Yt_l + ut (9) is also an AR(I) model.

• Following the procedure discussed for Random Walk Without Drift, it

can be shown that for the random walk with drift model (9),
E(Yt) ........... (1 1)
var (Yt ) — t02 . (12)
• Here, again for RWM with drift the mean as well as the variance
increases over time, again violating the conditions of stationarity.
• In short, RWM, with or without cliift, is a non-stationary stochastic
process.
• The random walk model is an example of what is known in the
literature as a Unit Root Process.

3) Unit Root Stochastic Process

• Let's mite the RWM Yt = Yt_l + (4) as:
+ ut_l -1 1 ....... (13)
• If p l, (13) becomes a RWM (Hithout drift).
• If p is in fact 1, we face what is known as the unit root problem
(non-stationarity); as the variance of Yt is not stationary.
• The name unit root is due to the fact that p = 1.
• Thus the terms non-stationaHty, random walk, and unit root can
be treated as synonymous.
• If, however, I pl I, then the time series Yt is stationary in the
sense we have defined it.

• Note: Unit Root Stochastic Process will be further explained in

Unit Root Test of Stationarity.
 Trend Stationary (TS) and Difference Stadonary (DS)

Deterministic Trend: if the trend in a time selies is completely predictable

• Stochasåc Trend: if it is not predictable

• E.g. consider the followmg model of the time series Yt . ßl + + ß3Yt_l +
ut......................(14)
• Now we have the following possibilities:
1. Pure Random Walk: If in (14) ßl = O, ß2 = O, m = I, we get:
Yt - + ............(15)
• Which is nothing but a RWM without drift and is therefore non-stationary.
But note that, if we "Tite (2.2) as
 Trend Stationary (TS) and Difference Stadonary (DS)
AYt = (Yt - = ............ (8)
• It becomes stationary, as noted before.
• Hence, a RWM without cliift is a Difference Stationary Process (DSP).

2. Random Walk With Drift: If in ßl + ß2t + ß3Yt_1 + lit...

-O, 1, we get:
Yt- ßl + Yt_l + ut ........(16)
• which is a random walk with drift and is therefore, non-stationary.
• If we MTite it as: = AYt = 131 + this means Yt
exhibit a positive (ßl > 0) or negative (ßl < 0) trend.
• Such a trend is called a Stochasdc Trend.
 Trend Stationary (TS) and Difference Stadonary (DS)

• Equation (16a) is a DSP process because the non-stationarity in Yt

can be eliminated by taking first differences of the time series.
3. Deterministic Trend: If in Yt— 131 + ß2t + ß3Yt_1 + ut ......
(14) ßl 0, ß2 + 0, m = 0, we obtain:
'31 + ß2t+ 1k ...........(17)
• Which is called a trend process (TSP).
• Although the mean of Yt is ßl + ß2t, which is not constant, its
variance ( = 02) is.
• Once the values of (31 and (32 are known, the mean can be
forecast perfectly.
 Trend Stationary (TS) and Difference Stadonary (DS)

• Therefore, if we subtract the mean of Yt from Yt, the resulting

series will be stationary, hence the name trend stationary.
4. Random Walk With Drift and Deterministic Trend: If in Yt = ßl
+ ß2t + ß3Yt_1 + lit ......(14) ßl * 0, 0, ß3 = l, we obtain: Yt = ßl
+ ß2t + Yt_l + ut ...... (18)
We have a random walk with drift and a deterministic trend, which
can be seen if we write this equation as:
AYt + ß2t + ut ....................(18a)
• Which means that Yt is non-stationary.
5. Determinisüc Trend With Stadonary AR(I) Component: If in
+ ß2t + + ut... (14) ßl + 0, * O, ß3 < 1, then we get:
Yt = ßl + ß2t + ß3Yt-l + ............(19)
 Trend Stationary (TS) and Difference Stadonary (DS)
• Which is stationary around the deterministic trend.
 The Phenomenon of Spurious Regression
Stationary Time Series are important, consider the following two
random walk models:

. (20) . (21)
• Where we generated 500 observations of ut from ut N(O, l) and
500 observations of Vt from Vt N(O, l) and assumed that the initial
values of both Y and X were zero.
• We also assumed that ut and Vt are serially uncorrelated as well as
mutually uncorrelated.
 The Phenomenon of Spurious
• Both these time series are non-stationary; i.e. they are I(l) or exhibit
stochastic trends.

Re ession (cont.)
Suppose we regress Yt on Xt
• Since Yt and Xt are uncorrelated 1(1) processes, the
R2 from the regression of Y on X should tend to zero;
that is, there should not be any relationship between
the two variables.
• But wait till you see the regression results:
Variable Coefficient Std. error t statistic
 The Phenomenon of Spurious
c -13 .2556 0.6203 -21 .36856 x 0.3376 0.0443 7 .61223

0.1044 d = 0 . 0121

• As you can see, the coefficient of X is highly statistically

significant, and, although the R2 value is low, it is statistically
significantly different from zero.
 The Phenomenon of Spurious
• From these results, you may be tempted to conclude that
there is a significant statistical relationship between Y and
X, whereas a priori there should be none.
• This is the phenomenon of spurious/non-sense regression,
first discovered by Yule.
 The Phenomenon of Spurious
• Yule showed that (spurious) correlation could persist
in non-stationary time series even if the sample is
very large.
• That there is something wrong in the preceding
regression is suggested by the extremely low
Durbin-Watson d value, which suggests very strong
first-order autocorrelation.
 Phenomenon of Spurious
R2 > d is a good rule of thumb to suspect that the
estimated regression is spurious, as in the example
above.
This example is a strong reminder that one should be
extremely cautious of conducting regression analysis
based on time series that exhibit stochastic trends.

Presented by Qazi Yasir Ayub

• In practice we face two important questions:
— How do we find out if a given time series is
stationary or not?
— Is there a way that it can be made stationary?
• Prominently discussed tests in the literature are:
— Graphical Analysis
• The Unit Root Test
• Before pursuing a formal test, it is always
advisable to plot the time series under study
• E.g. take the GDP time series.
• You will see that over the period of study GDP
has been Increasing (i.e. showing an upward
trend)
• This perhaps suggests that the GDP series is not
stationary (also more or less true of the other
economic time series).
 2- The Unit
• The widely popular test of stationarity is the unit root
test.

• We start with: Yt- + ut - 1 s p s l (1.1) Where ut is a white

noise error term.
• We know that if p 1 (i.e. in the case of the unit root) (l .
1) becomes a Random Walk Model Without Drift, (a
non-stationary stochastic process).
 2- The Unit Root Test (cont.)
• Therefore, why not simply regress Yton its lagged value
Yt_l and find out if the estimated p is statistically equal
to 1? If it is, then Yt is non-stationary.
• Procedure:
(1.1)
py I -Y&l+ut (4.1)
= Yt_l (p - l)+ ut or

which can be alternatively written as:

 2- The Unit Root Test (cont.)
(4.2) where
(p — 1) and A— first-difference (Yt — Y
• Hence, to estimate (4.2) and test the I-1(): 0.
• If 0, then p — I (i.e. unit root/time series under
consideration is non-stationary) and (21.9.2) will
become - (Yt- Yt_l) = (4.3)
• Since ut IS a white noise error term, It is stationary,
which means that the first differences of a random walk
time series are stationary.
• Now let's turn to the estimation of (4.2).
 2- The Unit Root Test (cont.)
• We have to take the first differences of Yt and regress them
on Yt_l and see if the estimated slope co-efficient in this
regression (6) is zero or not.
• If it is zero, we conclude that Yt is non-stationary.
• But if it is negative, we conclude that Yt is stationary.
• The only question is which test we use to find out if the
estimated co-efficient of Yt_l in (4.2) is zero or not?
• Unfortunately, under the null hypothesis that = 0 (i.e., p
l), the t value of the estimated coefficient of Yt_l does not
follow the t distribuüon even in large samples; i.e. it does
not have an asymptotic normal distribution.
 the
• Dickey and Fuller have shown that under the null
hypothesis that ö 0, the estimated t-value of the
coefficient of Yt_l in (4.2) follows the T (tau)
statistic.
• These authors have computed the critical values
of the tau staüsüc on the basis of Monte Carlo
simulations.
• Interestingly, if the hypothesis that = 0 is rejected
(i.e. the time series is stationary), we can use the
usual t test.
• The actual procedure of implementing the DF test
involves several decisions.
• In discussing the nature of the unit root process in
Sections 21.4 and 21.5, we noted that a random
walk process may have no drift, or it may have drift
or it may have both deterministic and stochastic
trends.
• To allow for the various possibilities, the DF test
is estimated in three different forms, i.e. under
three different null hypotheses.
• Yt is a random walk: Yt 5Yt_l + ut (4.2)
• Yt is a random walk with drift: Yt ßl + 5Yt_l + ut (4.4)
• Yt is a random walk with drift around a stochastic
trend: = ßl + ß2t + 6Yt_l + (4.5)
Where t is the time or trend variable.
• In each case, the null hypothesis is that Ö 0; i.e.
there is a unit root—the time series is nonstationary.
• The alternative hypothesis is that is less than zero;
i.e. the time series is stationary.
• If the null hypothesis is rejected, it means that Yt is
a stationary time senes with zero mean in the case
of (4.2), that Yt is stationary with a nonzero mean
ßl/(l — p)] in the case of (4.4), and that Yt is
stationary around a deterministic trend in (4.5).
• It is extremely important to note that the critical values of
the tau test to test the hypothesis that = 0, are different for
each of the preceding three specifications of the DF test,
which can be seen clearly from Appendix D, Table D. 7.
• Moreover, if, say, specification (4.4) is correct, but we
estimate (4.2), we will be committing a specification error,
 whose consequences we already know from Chapter 13.
The same is frue if we estimate (4.4) rather than the true
(4.5).
• Of course, there is no way of which specification is correct to
begin with.
• Some trial and error is inevitable, data mining
notwithstanding.
• The actual estimation procedure is as follows:
• Estimate (4.2), or (4.3), or (4.4) by OLS; divide the estimated
coefficient of Yt—l in each case by its standard error to
 compute the (T) tau statistic; and refer to the DF tables (or any
statistical package).
• If the computed absolute value of the tau statistic (IT ) exceeds
the DF or MacKinnon critical tau values, we reject the
hypothesis that 0, in which case the time series is stationary.
• On the other hand, if the computed IT I does not exceed the
critical tau value, we do not reject the null hypothesis, m
which case the time senes is non-stationary.
• Make sure that you use the appropriate critical T values.
• Let us return to the U.S. GDP time series. For
this series, the results of the three regressions
(4.2), (4.4), and (4.5) are as follows:
• The dependent variable in each case is Yt
GDPt
• GDPt- O.00576GDPt-ı...... (4,6)
• t- (5.7980) 112 --0.0152 d- 1.34
• GDPt -z 28.2054 - O.00136GDPt-ı. (4.7) t-
(1.1576) (-0.2191) - O.00056 d 1.35
• GDPt- 190.3857 + 1.4776t- O.0603GDPt-ı
(1.8389) (1.6109) (-1.6252)........
(4,8)

• 112 - d- 1.31
• Our primary interest here is in the t ( = t) value of
the GDPt-1 coefficient.
• The critical 1, 5, and 10 percent T values for
model (4.6) are —2.5897, -1.9439, and -
1.6177, respectively, and are —3.5064, -
2.8947, and -2.5842 for model (4.7) and -
4.0661, -3.4614, and —3.1567 for model
(21.3.8).
• As noted before, these critical values are different
for the three models.
• Before we examine the results, we have to decide
which of the three models may be appropriate.
• We should rule out model (4.6) because the
coefficient of GDPt—l, which is equal to 6 is
positive.
• But since Ö (p — l), a positive ö would imply that
1.
• Although a theoretical possibility, we rule this case out
because in this case the GDP time series would be
explosive.
• That leaves us with models (4.7) and (4.8). In both
cases the estimated coefficient is negative, implying
that the estimated p is less than 1.
• For these two models, the estimated p values are
0.9986 and 0.9397, respectively.
• The only question now is if these values are
statistically significantly below I for us to declare that
the GDP time series is stationary.
• For model (4.7) the estimated T value is —0.2191, which
in absolute value is below even the 10 percent critical value
of -2.5842.
• Since, in absolute terms, the former is smaller than the
latter, our conclusion is that the GDP time series is not
stationary.
• The story is the same for model (4.8).
• The computed t value of —1.6252 is less than even the 10
percent critical T value of —3.1567 in absolute terms.
• Therefore, on the basis of graphical analysis, the
correlogram, and the Dickey-Fuller test, the conclusion is
that for the quarterly periods of 1970 to 1991, the U.S.
GDP time series was non-stationary; i.e., it contained a unit
root.