© Bank of England VARs, SVARs and VECMs Ibrahim Stevens Joint HKIMR/CCBS Workshop Advanced Modelling for Monetary Policy in the Asia-Pacific Region May 2004
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 1/57
© Bank of England
VARs, SVARs and VECMs
Ibrahim Stevens
Joint HKIMR/CCBS Workshop
Advanced Modelling for Monetary Policy inthe Asia-Pacific Region
May 2004
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 2/57
© Bank of England
Contents
• What is a VAR ?
• Advantages and disadvantages of VARs
• Number of lags
• Identification
• Impulse response functions
• Variance decomposition• SVARs
• VECMs
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 3/57
© Bank of England
What is a VAR?
•
Every stationary univariate process has anautoregressive representation (see Wold
decomposition):
• Useful if the process can be modelled well using few
lags, eg
• must be small relative to xt and be „well- behaved‟
(eg white noise)
......2211
nt nt t t x x x x
t t t t x x x 2211
t
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 4/57
© Bank of England
What is a VAR?
•
Let {x t } be a sequence of vectors rather than scalarsand you have a VAR!
• The B’ s are now matrices of coefficients and is a
vector of constants. (It is useful to work with few lags)
•Again we wish to be small relative to x t and to be
„well- behaved‟ (eg white noise)
......2211
nt nt t t x B x B x B x
t t t t x B x B x 2211
t
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 5/57
© Bank of England
Why use VARs
• VARs treat all variables in the system as
endogenous
• (Might) avoid „incredible identification‟ problems• Model the dynamic response to shocks
• Can aid identification of shocks - including
monetary policy shocks• Often considered good for short-term forecasting
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 6/57
© Bank of England
Problems with using VARs
• Identification is a big issue
– different identification schemes can give very
different results – ditto different lag lengths
• Tend to be over-fitted (too many variables and lags)
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 7/57
© Bank of England
An Example Of Economic Model
• Firms set free prices ( f ) on the basis of current overall
price levels;
• Government sets administered prices ( g ) on the basis
of current overall price level;
• The overall price level is a weighted sum of free and
administered pricesSo:
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 8/57
© Bank of England
Structural system of interest
t t t t t f b f a g b g 111212111
t t t t t f b g b g a f 212212121
t t t p B p A
1
Or,
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 9/57
© Bank of England
t t t p B p A
1
t
t
t
t
t
t f g p
bbbb B
aa
2
1
2221
1211
21
12 , , ,1
1
Or,
t t L D p LC )()(
Structural system of interest
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 10/57
© Bank of England
Where,
t t L D p LC )()(
Structural system of interest
I L D BL A LC )( ,)(
Are polynomials in the lag operator L
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 11/57
© Bank of England
Big Problem 1
We can‟t estimate the structural system directly
Instead of:
We estimate the reduced form:t t t
p B p A 1
t t t A p B A p 1
1
1
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 12/57
© Bank of England
Estimation issue 1
• Which estimation method should we choose?
• We can use OLS because all variables are the same
on the RHS for each equation
• As long as we use the same number of lags for each
equation
• [If not, more efficient to use SUR] – Near-VAR models
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 13/57
© Bank of England
Estimation issue 2
• Should the variables in the VAR be stationary?
• Non-stationary variables might lead to spurious results
• But Chris Sims argues that, by differencing or
detrending, a lot of information is lost
• And that we are interested in the inter-relationships
between variables, not the coefficients themselves
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 14/57
© Bank of England
Estimation issue 3
• How many lags should we use?
• More lags improve the fit …
• … but reduce the degrees of freedom and increase the
likelihood of over-fitting
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 15/57
© Bank of England
Overfitting
•
For a stationary process
• But this does not imply that you improve the model
by increasing the number of lags
• A big problem is noisy data. By increasing the
number of explanatory variables you may be
„explaining‟ noise, not the underlying DGP
0limwhere,n
1
t t
n
iit it
x x
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 16/57
© Bank of England
Estimation issue 3
Two ways round this problem:
• Economic theory might suggest appropriate lag
length
• Use lag length selection criteria: trade-off parsimony
against fit
• Think about the results
– do you believe them?
– do they suggest you should seasonally adjust?
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 17/57
© Bank of England
Big Problem 2
• We can move easily from the structural model to the
reduced form
• But not the other way
• In this example we estimate 7 parameters but the
structural model has 8
• Therefore there are an infinite number of ways of
moving from reduced form to the unobservedstructural model
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 18/57
© Bank of England
Big Problem 2
• So, 8 unknowns and 7 estimates, hmm
• We need to make one of the unknowns known
• ie we need to set the value of (at least) one coefficient
[make an identifying restriction]
• If we make 1 identifying restriction the system is just
identified
• If we make more than 1, it is over identified: we canthen test restrictions
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 19/57
© Bank of England
Unrestricted VARs and SVARs
To summarise:
• We estimate an unrestricted (or reduced-form) VAR
• By imposing suitable theoretical restrictions we can
recover the restricted (or structural) VAR
• The unrestricted VAR is a statistical description of the
data, the SVAR adds some economics
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 20/57
© Bank of England
Unrestricted VARs vs SVARs
• Unrestricted VAR compatible with a lot of theories:
produces good short-term forecast
• Structural VAR tied to a particular theory: provides a
better interpretation of forecast.
• Long-term forecast is a combination of both
• Structural VAR is better for a Central Bank??
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 21/57
© Bank of England
Identifying restrictions
1. Recursive/Wold chain/Cholesky decomposition.
Sims‟ lag of data availability
2. Coefficient restriction: eg a12=1
3. Variance restriction: eg Var(1)=3
4. Symmetry restriction : eg a12=a21
5. Long-run restrictions, eg nominal shocks don‟t affectreal variables.
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 22/57
© Bank of England
Example
The g equation of,
is,
t t t A p B A p 1
1
1
]
[1
1
12121122212
1211211
2112
t t t
t t
a f bba
g babaa g
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 23/57
© Bank of England
Example
If a12 =0 we have,
ie the reduced-form error on the equation is the same as
the structural error
• This is the Cholesky decomposition
• We are assuming a recursive ordering of the
transmission of fundamental shocks in the system
t t t t f b g b g 1112111
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 24/57
© Bank of England
Cholesky decomposition
In order to recover the fundamental shocks to the system
we restrict the A matrix to be triangular (zeroes above or
below the main diagonal)
• We need to be confident about the ordering of the
impact of shocks
• The frequency of the data becomes a big issue• Nonetheless, probably the most common form of
identification
t t t p B p A
1
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 25/57
© Bank of England
Cholesky Decomposition
Take the system:
Then:
M
t
Y
t
t
t
t
t
v
v B
M
Y LC
M
Y A
1
1)(
nn
ii
nnn
b
b
b
B
aa
a A
00
00
00
1.
.1..001
000111
11
21
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 26/57
© Bank of England
Monetary Policy
“There is little hope that economists can evaluate
alternative theories of monetary policy transmission, or
obtain quantitative estimates of the impact of monetary policy changes on various sectors in the economy, if there
exists no reasonably objective means of determining the
direction and size of changes in policy stance”
Bernanke and Mihov (QJE , 1998)
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 27/57
© Bank of England
Using the Cholesky decomposition
• Huge literature on identifying monetary policy shocks
• Typically use recursive identification schemes
• Note that if we are interested in just one of the structuralshocks we don‟t have to identify the whole system
• Nonetheless, controversies abound
– eg what is the monetary policy instrument
– what is the information set of the policy maker
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 28/57
© Bank of England
Assessing the stance of MP
• Bernanke and Mihov (1998) use an SVAR to determine
the stance of monetary policy
• Essentially they estimate an MCI but try to get round
the problem of shock identification
• They claim they do this by identifying fundamental
shocks to policy
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 29/57
© Bank of England
Bernanke and Mihov - method
• They use a block -recursive identification scheme with
two blocks: policy and non-policy
• They assume that policy has no effect on the non-policyvariables in the initial period
• They identify shocks to the policy variables (FFR,
exchange rate, term spread)
• But do not identify the fundamental shocks to the non-
policy part of the system
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 30/57
© Bank of England
Bernanke and Mihov – semi-structural
VARs
• Take the VAR with variables GDP (Y), CPI (P), price
of raw materials (Pcm), Federal Funds Rate (FF), total
bank reserves (TR) and non-borrowed reserves (NBR)• They assume:
1. Orthogonality of structural disturbances
2. Macroeconomic variables do not react to changes in
the monetary variables
3. Restrictions on the monetary block
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 31/57
© Bank of England
Bernanke and Mihov – semi-structural
VAR
For point 3, Bernanke y Mihov assume:
S B B D D NBR
B FF BR
D FF TR
vvvu
vuu
vuu
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 32/57
© Bank of England
Bernanke and Mihov – semi-structural VAR
To identify the structural disturbances:
S
D
B
NP NP
NP
D B
NBR
TR
FF
Pcm
P
Y
v
v
v
v
v
v
u
u
u
u
u
u
a
a
a
a
a
a aa
a
a
a
a
3
2
1
11
11
63
53
62
52
61
514342
32
41
31
21
1
0
00
0
1
0
00
0
0
0
0
0
0
0
0
0
0
1
0
0
0
00010
0001
1
0
00
0
0
1
00
0
0
1
01
001
0001
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 33/57
© Bank of England
Bernanke and Mihov – semi-structural VAR
• One still needs to make other restrictions which
depend on theory
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 34/57
© Bank of England
Blanchard-Quah
• Alternative way of identifying the VAR is to think
about variables interaction in the long run
• Blanchard and Quah (AER, 1989) advocate an
identification scheme where
– supply shocks have permanent effects on real
variables
– demand shocks only have temporary effects on realvariables
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 35/57
© Bank of England
Blanchard-Quah • Take the structural VAR we are interested in (
Blanchard-Quah use GDP and unemployment)
•As a VMA,
10
01
)var(),cov(
),cov()var(
,,)( 1
t u
t u
t y
t u
t y
t y
u
y z B z L A z
t u
t y
t t
t t t t t
t t B L A z 1)(1
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 36/57
© Bank of England
Blanchard-Quah
• The VAR we estimate (reduced form) in VMA
form
•We must impose restrictions on D(L) to obtain
A(L) and B. Assume (1-A(L))-1B=C(L)
t t e L D z )(
)()()()()(,)(
2221
1211
LC LC LC LC LC LC z t t
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 37/57
© Bank of England
Blanchard-Quah • We can see
• If we assume that shocks to GDP are demand shocks and
shocks to unemployment are supply shocks, we can
assume that GDP shocks will not have a permanent effect
on GDP, that isC 11(L) yt =0
• Under this assumption one can recover the demand and
supply shocks from the reduced VAR
t u
t y
t
t
LC LC
LC LC
u
y
)()(
)()(
2221
1211
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 38/57
© Bank of England
Using Blanchard-Quah
• Quah and Vahey ( EJ , 1995) use the BQ technique to
identify core inflation
• They define core inflation as that part of inflation that
has persistent effects on the price level
• So core inflation is like a real variable in the BQ paper
and transitory inflation like a nominal variable
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 39/57
© Bank of England
Faust and Rogers (2000)
• Investigate: – How delayed is the overshooting of the exchange
rate following a nominal shock
– Test to see how well UIP fares
– Ask what proportion of the volatility of the
exchange rate can be explained by MP shocks
• Find:
– delay of overshooting is sensitive to assumptions
– UIP performs miserably
– between 2 and 30% of the volatility is explained by
MP shocks
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 40/57
© Bank of England
Faust and Rogers - Technique
• Under identify the VAR arguing:
– few variables make identification easier but you have
omitted variables
– but you can only be confident about a fewrestrictions with bigger systems
• Then run the VARs using other possible restrictions
• If the impulse responses are similar: fantastic!
• Otherwise choose the impulse response you think best
fits reality
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 41/57
© Bank of England
Impulse Response Functions
• Perhaps the most useful product of VARs
• Use the moving average representation of a VAR to
show the dynamic response of variables to fundamentalshocks
• Often used to determine the lags of the monetary
transmission mechanism, etc
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 42/57
© Bank of England
Example of going from AR to MA
Take a VAR with one lag
n A
A x A x
A x A x
Ax x
Ax x
i it
i
n
i it
i
nt
n
t
t t t t
t t t
t t t
as
....
0
1
0
12
2
121
1
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 43/57
© Bank of England
Impulse response for an AR(1)
• The effect on { xt } of a one-period shock is given by the
MA representation
• In this case• Here
0 t
t x
-1.2
-0.8
-0.4
0
0.4
0.8
1.2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
8.0
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 44/57
© Bank of England
Variance decomposition
• Analyses the relative „importance‟ of variables
• More precisely, the proportion over time of the variance
of a variable due to each fundamental shocks
• eg, in the Faust and Rogers paper, they look at what
proportion of the volatility of the exchange rate is
explained by the various shocks
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 45/57
© Bank of England
General principles of identifying
VARs
• Identify what you are interested in
– eg if you simply want a forecast, do you need to
identify it at all?
• Think about the economics
• Test for robustness
• Compare impulse responses - do you believe them?
– eg the „price puzzle‟
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 46/57
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 47/57
© Bank of England
VECMs
• Vector error correction mechanisms are generalisationsof ECMs
ECM:
VECM:
z t are the equilibrium relationships
111 t t t t t x y y x y
11)( t t t Cy y L B y A
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 48/57
© Bank of England
VECMs
• Like ECMs, VECMs combine short-run (dynamic)
information with long-run (static) information
• They are like a combination of an unrestricted VAR (the
dynamic part) and a structural VAR (the long-run is(should be) consistent with theory)
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 49/57
© Bank of England
Identification Issues
• As with VARs you estimate it with no current
dependent variables on the RHS
• But theory might suggest that the structural model
should include them• Finding the cointegrating vectors (and even knowing
how many there are) is a fragile business
• If there are n variables in the system there are (n-1) possible cointegrating vectors
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 50/57
© Bank of England
Johansen
• You are testing the rank of the matrix A-1C in the
equation below
• If it has less than full rank there is at least one
cointegrating vector
1
1)(
t t t Cy A y L D y
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 51/57
© Bank of England
Johansen with 2 variables
• If there is an equilibrium relationship between x and z
• The are the „loadings‟ and gives the equilibrium
relationship
1
1
2
1
1
1
1 t
t
t z
x
Cy A
s'
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 52/57
© Bank of England
Estimating a VECM: technicalities
• Lag length of VECM: - Gaussian residuals
• Worth considering exogenous short-run variables (eg
dummies, other economic variables)
• Inclusion of deterministic components (constant, trend)
- Johansen (1992) suggests:
- Begin with estimation of most restrictive model (no
constant, no trends) and estimate less and less
restrictive variants until reach least restrictive(constant, trends)
- Compare trace stats to their critical level
-Stop the first time null is not rejected
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 53/57
© Bank of England
Estimating VECM: technicalities
•Identification of cointegrating vectors:
- Johansen test tells you the number of cointegrating
vectors; but it does NOT tell you if these are unique
- If one cointegrating vector, then the vector isidentified
- If more than one cointegrating vector, we must
impose restrictions
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 54/57
© Bank of England
Estimating VECM: technicalities
•Imposing restrictions is „easy‟. These can be imposedas follows: There are r cointegrating vectors (found in
matrix ), then
H : =(H11, H22,…,Hr r )
where Hi i=1,…,r are matrices representing the linear
economic relationships to be tested and i i=1,…,r are
vectors with parameters to be estimated (for each of the
cointegrating relationships)
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 55/57
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 56/57
© Bank of England
Gonzalo and Ng (2001)
•
The „loadings‟ identify the effects of the disequilibrium
terms on the endogenous variables
• Show that the orthogonal complement of loadings and
cointegrating vectors:
can be used to define Permanent and Transitory shocks• Apply Choletsky, can do variance decomposition
'
'
G
8/11/2019 VAR, SVAR and VECM
http://slidepdf.com/reader/full/var-svar-and-vecm 57/57
© B k f E l d
Textbooks
• Hamilton, James D, (1994) „Time Series Analysis‟, Princeton University Press
• Enders, Walter, (1995) „Applied econometric time series‟, Wiley • Harris, Richard (1995), „Cointegration Analysis in Econometric Modelling‟, Harvester Wheatsheaf
Identifying Monetary Policy Shocks
• Christiano, Lawrence J.; Eichenbaum, Martin; Evans, Charles L, (1999) „Monetary Policy Shocks:
What Have We Learned and to What End?‟ in Taylor, John B.; Woodford, Michael, eds. Handbook
of Macroeconomics. Volume 1A. Handbooks in Economics, vol. 15, North Holland
• Leeper, Eric M.; Sims, Christopher A.; Zha, Tao, (1996) „What Does Monetary Policy Do?‟,
Brookings Papers on Economic Activity; 0(2), 1996, pages 1 63
• Bernanke, Ben S.; Mihov, Ilian (1998) „Measuring Monetary Policy‟, Quarterly Journal of
Economics; 113(3), August 1998, pages 869 902.
Blanchard Quah
• Blanchard, Olivier Jean; Quah, Danny. (1989) „The Dynamic Effects of Aggregate Demand andSupply Disturbances‟, American Economic Review; 79(4), September 1989, pages 655 73
• Quah, Danny; Vahey, Shaun P (1995) „Measuring Core Inflation‟, Economic Journal; 105(432),
September 1995, pages 1130 44
Others
• Faust, John and John Rogers, (2003?), „Monetary Policy‟s Role in Exchange Rate Behaviour‟,