Scribd
Upload
65 views9 pages

Granger Slides

Uploaded by

okoyejohnpaul48
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
65 views9 pages

Granger Slides

Uploaded by

okoyejohnpaul48
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/ 9

ECO 431

TIME SERIES ECONOMETRICS

BY

DR. OBODOECHI DIVINE N


1
Granger Causality
• In regression, we label one variable the dependent variable and the others the
explanatory variables.
• In many cases, because the latter “explained” the former it was reasonable to
talk about X “causing” Y.
• If event A happens before event B, then it is possible that A is causing B.
However, it is not possible that B is causing A.
• These ideas can be investigated through regression models using the notion of
Granger causality.
• it is important to note that GC does not mean causation rather, it indicates that
there is a correlation btw the past values of one variable and the present value
of another.
• X “Granger causes” Y if past values of X can help explain Y.
• If Granger causality holds this does not guarantee that X causes Y. But, it
suggests that X might be causing Y.
• Granger causality is only relevant with time series variables.

2
Granger Causality in a Simple ARDL Model
• Since X and Y are assumed to be stationary, use the ADL model.
• Begin with simple ARDL(1,1) model:

• β1 is a measure of the influence of Xt-1 on Yt. If β1=0 then X does not


Granger cause Y.

• “if β1=0 then past values of X have no explanatory power for Y


beyond that provided by past values of Y”.

• Alternatively, If β1 is statistically significant (e.g. P-value < .05) then


we conclude that X Granger causes Y.

3
Granger Causality in an ARDL Model with p and q Lags

• Granger Causality with many lags in an ARDL Model can be


represented as thus;

• X Granger causes Y if any or all of β1, ...,βq are statistically


significant.
• The proper way to do Granger causality testing is to test the
hypothesis that β1=β2=...=βq=0
• X Granger causes Y only if the hypothesis is rejected.

• Note that the joint test of β1=β2=...=βq=0 is not exactly the same as q4
Granger Causality in an ARDL Model with p and q Lags

• Sometimes researchers check for Granger causality simply (albeit


imperfectly) using only t-tests.
• The P-values for the t-states on individual coefficients can be used to
determine whether Granger causality is present.
• Using the 5% level of significance, then if any of the P-values for the
coefficients β1,...,βq were less than .05, you would conclude that
Granger causality is present.
• If none of the P-values is less than .05 then you would conclude that
Granger causality is not present.

5
Granger Causality in an ARDL Model with p and q Lags

• Warning if you use t-tests:

• If any or all of the coefficients β1,...,βq are significant using t-tests, you may safely
conclude that X Granger causes Y.

• If none of these coefficients is significant, it is probably the case that X does not Granger
cause Y. However, you are more likely to be wrong if you conclude the latter than if you
had used the correct joint test of Granger non-causality.

• Example

• We have quarterly data on construction share and public debt for Nigeria.

• ADF and Johansen tests indicate that both CS and PD appear to have unit roots, but are
not cointegrated.

• However, differences of these series are stationary, you will have to use the differenced
variables of those series to investigate whether CS in Nigeria Granger causes PD.

6
Direction of Causality
• The direction of causality can be either Unidirectional, or
Bidirectional.
• The direction of causality is determined by the significance of the p-
values in our results.

• when two series A and B are tested for causality, and it is found that
series A GCs series B and series B does not GCs A, then the direction
of causality is said to be Unidirectional.

• On the other hand, when two series A and B are tested for causality,
and it is found that series A GCs series B and series B also GCs A,
then the direction of causality is said to be Bidirectional.

7
Estimating Granger Causality for Two Series
• When estimating GC for two series, it is important to state the null
hypothesis and also the decision rule in other to determine the
direction of causality.
• GC is conducted on the stationary series and hence the two series
involed must be stationary (I(1)) in order to conduct the test.
• Stating the null hyothesis
• H0: No causal relationship (no causality)
• H1: Causal relationship (causality)
• Decision Rule
• Reject H0 when the p-value at the 5% LS is less than 0.05 and then
conclude that series A GCs B, hence a unidirectional causality running
from series A to B.
• furthermore, when the H0 is rejected for both series A and B then there
is Bidirectional causality running from both series A and B.
• Lets look at practical examples on Eviews

8
THANK YOU!

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy