Scribd
Upload
61 views

BEC 3350/4340:ECONOMETRICS: Session 8 Stationarity of A Time Series

The document discusses stationarity of time series data. It defines a stationary stochastic process as one where the mean and variance are constant over time. Nonstationary processes include random walks, which can be with or without drift. A unit root process is also considered nonstationary. For a time series to be useful for forecasting, it needs to be stationary. The document discusses different types of stochastic processes that can cause nonstationarity, such as difference stationary processes, trend stationary processes, and integrated processes of varying orders. Finally, it lists some common tests used to test for stationarity, including scatter diagrams, autocorrelation functions, and unit root tests.

Uploaded by

Nuwani Manasinghe
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
61 views

BEC 3350/4340:ECONOMETRICS: Session 8 Stationarity of A Time Series

The document discusses stationarity of time series data. It defines a stationary stochastic process as one where the mean and variance are constant over time. Nonstationary processes include random walks, which can be with or without drift. A unit root process is also considered nonstationary. For a time series to be useful for forecasting, it needs to be stationary. The document discusses different types of stochastic processes that can cause nonstationarity, such as difference stationary processes, trend stationary processes, and integrated processes of varying orders. Finally, it lists some common tests used to test for stationarity, including scatter diagrams, autocorrelation functions, and unit root tests.

Uploaded by

Nuwani Manasinghe
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

BEC 3350/4340:ECONOMETRICS

Session 8
Stationarity of a Time Series

Amanda Perera
Lecturer
Department of Business Economics
 Introduction to time series data
Lecture  Stationarity of time series data
Outline  Consequences of using nonstationary data
 Testing stationarity
1. Stochastic Processes
A random or stochastic process is a collection of random
variables ordered in time

Concepts:
2. Stationary Stochastic Processes
A stochastic process is said to be stationary if its mean
and variance are constant over time
A purely random, or white noise, process is process where it
has zero mean, constant variance σ2, and is serially uncorrelated
Why are stationary time series so important?
Because if a time series is nonstationary, we can study its
behavior only for the time period under consideration.
Each set of time series data will therefore be for a
particular episode. As a consequence, it is not possible to
generalize it to other time periods. Therefore, for the
purpose of forecasting, such (nonstationary) time series
may be of little practical value.
Random Walk
3. Model (RWM)
Nonstationary
Stochastic
Processes Random walk Random walk
without drift with drift
A random walk without drift A random walk with drift
 𝒀 𝒕 =𝜶 𝒀 𝒕 −𝟏 +𝒖𝒕

 If is in fact 1, we face what is known as the unit root


problem, that is, a situation of nonstationarity; we
4. Unit Root already know that in this case the variance of Yt is not
Stochastic  stationary.
 
 Thus the terms nonstationarity, random walk, unit
Process root, and stochastic trend can be treated
synonymously.
 If the trend in a time series is a deterministic function of time, such
as time, time-squared etc., we call it a deterministic trend, whereas
if it is not predictable, we call it a stochastic trend. To make the
definition more formal, consider the following model of the time
series Yt .

5.Difference Stationary  If , , we get pure random walk (RWM without drift) which is
(DS) Processes  nonstationary
 
  𝒀 𝒕 =𝒀 𝒕 −𝟏 + 𝒖𝒕
 But note that, if we write the Equation as

∆
  𝒀 𝒕=(𝒀 ¿ ¿ 𝒕 − 𝒀 𝒕 − 𝟏)= 𝒖𝒕 ¿

it becomes stationary, as noted before. Hence, a RWM without


drift is a difference stationary process (DS).
 If , , we get Random walk with drift and is therefore nonstationary

 
 If we write it as;

 this means will exhibit a positive ( β 1 > 0) or negative ( β 1 < 0)


trend. Such a trend is called a stochastic trend.
 If , , we get
6. Trend
Stationary  

(TS)Process
 which is called a trend stationary process (TS).
 If a time series has to be differenced once (first difference of
the series) , to make it stationary, we call such a time series
integrated of order 1. Similarly, if a time series has to be
differenced twice (i.e., take the first difference of the first
differences) to make it stationary, we call such a time series
integrated of order 2.
7. Integrated  In general, if a (nonstationary) time series has to be
Stochastic differenced d times to make it stationary, that time series is
said to be integrated of order d. A time series Yt integrated
Processes of order d is denoted as Yt ∼ I (d). If a time series Yt is
stationary to begin with (i.e., it does not require any
differencing), it is said to be integrated of order zero,
denoted by Yt ∼ I (0). Thus, we will use the terms “stationary
time series” and “time series integrated of order zero” to
mean the same thing.
1. Scatter diagram

Tests of
2. Autocorrelation Function (ACF)
Stationarity
3. The Unit Root Test
Thank You…

You might also like