Assessing structure in monetary policy models Ragnar Nymoen Model Evaluation in Macroeconomics

advertisement
Assessing structure in monetary policy models
Ragnar Nymoen
http://folk.uio.no/rnymoen/
Department of Economics
University of Oslo
Model Evaluation in Macroeconomics
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
1 / 19
Contents
1
Introduction
2
Structural interpretation and structural properties
3
Evaluation of the New Keynesian Phillips Curve
4
Evaluation of an econometric model of the inflation spiral
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
2 / 19
Contents
1
Introduction
2
Structural interpretation and structural properties
3
Evaluation of the New Keynesian Phillips Curve
4
Evaluation of an econometric model of the inflation spiral
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
2 / 19
Contents
1
Introduction
2
Structural interpretation and structural properties
3
Evaluation of the New Keynesian Phillips Curve
4
Evaluation of an econometric model of the inflation spiral
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
2 / 19
Contents
1
Introduction
2
Structural interpretation and structural properties
3
Evaluation of the New Keynesian Phillips Curve
4
Evaluation of an econometric model of the inflation spiral
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
2 / 19
Introduction
Structural models
Structure has a positive ring about it, so several model types have
been dubbed structural
Simultaneous equations models
Ditto, but dynamic
Large scale econometric models
Structural VARs
Models with ’deep structural’ parameters
In modern macroeconomics, after Lucas’ reorientation, structural
model has come to mean ’derived from modern Walrasian theory’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
3 / 19
Introduction
Structural models
Structure has a positive ring about it, so several model types have
been dubbed structural
Simultaneous equations models
Ditto, but dynamic
Large scale econometric models
Structural VARs
Models with ’deep structural’ parameters
In modern macroeconomics, after Lucas’ reorientation, structural
model has come to mean ’derived from modern Walrasian theory’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
3 / 19
Introduction
Structural models
Structure has a positive ring about it, so several model types have
been dubbed structural
Simultaneous equations models
Ditto, but dynamic
Large scale econometric models
Structural VARs
Models with ’deep structural’ parameters
In modern macroeconomics, after Lucas’ reorientation, structural
model has come to mean ’derived from modern Walrasian theory’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
3 / 19
Introduction
Structural models
Structure has a positive ring about it, so several model types have
been dubbed structural
Simultaneous equations models
Ditto, but dynamic
Large scale econometric models
Structural VARs
Models with ’deep structural’ parameters
In modern macroeconomics, after Lucas’ reorientation, structural
model has come to mean ’derived from modern Walrasian theory’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
3 / 19
Introduction
Structural models
Structure has a positive ring about it, so several model types have
been dubbed structural
Simultaneous equations models
Ditto, but dynamic
Large scale econometric models
Structural VARs
Models with ’deep structural’ parameters
In modern macroeconomics, after Lucas’ reorientation, structural
model has come to mean ’derived from modern Walrasian theory’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
3 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Theoretical underpinnings imply structure?
OK, if purpose is to analyze counterfactual observations from a
theoretical system
but requires evaluation when the model is used to explain
observations of the macro economy (e.g., last and next years
national accounts and inflation),
and for decision making
In practice the usefulness of a model hinges on other properties
as well, for example
I
I
I
explanatory power (’fit)
stability (over time) and constancy (across regimes)
encompassing of other evidence
Important to evaluate also these other dimensions.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
4 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Structural properties
Several dimensions: theoretical interpretation, fit, constancy,
encompassing. )
When all features are present, a model has achieved a structural
representation of the macroeconomy.
In practice, structure is partial:
I
I
all models are wrong; and are better suited for some purposes than
others
all models break down sooner or later
Nevertheless better to know your model’s limitations than to
postulate that it mimics reality to perfection (or better than other
models).
In practice also useful to know which part of a model is relatively
structural, and which is ’soft as a grape’.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
5 / 19
Introduction
Layout of the paper
Structural model (rhetorics and substance).
Evaluation example I: The New Keynesian Phillips Curve
Can structural representations of the expected to emerge form
Walrasian microfoundations?
Back to square one: A Marshallian approach to macro-modelling
Evaluation example II: An econometric model of the inflation-spiral
(price setting wages and wage setting)
Summary and discussion
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
6 / 19
Introduction
Layout of the paper
Structural model (rhetorics and substance).
Evaluation example I: The New Keynesian Phillips Curve
Can structural representations of the expected to emerge form
Walrasian microfoundations?
Back to square one: A Marshallian approach to macro-modelling
Evaluation example II: An econometric model of the inflation-spiral
(price setting wages and wage setting)
Summary and discussion
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
6 / 19
Introduction
Layout of the paper
Structural model (rhetorics and substance).
Evaluation example I: The New Keynesian Phillips Curve
Can structural representations of the expected to emerge form
Walrasian microfoundations?
Back to square one: A Marshallian approach to macro-modelling
Evaluation example II: An econometric model of the inflation-spiral
(price setting wages and wage setting)
Summary and discussion
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
6 / 19
Introduction
Layout of the paper
Structural model (rhetorics and substance).
Evaluation example I: The New Keynesian Phillips Curve
Can structural representations of the expected to emerge form
Walrasian microfoundations?
Back to square one: A Marshallian approach to macro-modelling
Evaluation example II: An econometric model of the inflation-spiral
(price setting wages and wage setting)
Summary and discussion
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
6 / 19
Introduction
Layout of the paper
Structural model (rhetorics and substance).
Evaluation example I: The New Keynesian Phillips Curve
Can structural representations of the expected to emerge form
Walrasian microfoundations?
Back to square one: A Marshallian approach to macro-modelling
Evaluation example II: An econometric model of the inflation-spiral
(price setting wages and wage setting)
Summary and discussion
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
6 / 19
Introduction
Layout of the paper
Structural model (rhetorics and substance).
Evaluation example I: The New Keynesian Phillips Curve
Can structural representations of the expected to emerge form
Walrasian microfoundations?
Back to square one: A Marshallian approach to macro-modelling
Evaluation example II: An econometric model of the inflation-spiral
(price setting wages and wage setting)
Summary and discussion
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
6 / 19
Structural interpretation and structural properties
Structural dimensions
Claims of structural representation of the macroeconomy can be
evaluated along several dimension, for example
I
I
I
I
Theoretical interpretation
Ability to explain data
Ability to explain earlier findings (properties of existing models)
Robustness to new evidence (sample period extension) and new
economic analysis (hypotheses of explanatory variables)
Pagan Frontier
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
7 / 19
Structural interpretation and structural properties
Structural dimensions
Claims of structural representation of the macroeconomy can be
evaluated along several dimension, for example
I
I
I
I
Theoretical interpretation
Ability to explain data
Ability to explain earlier findings (properties of existing models)
Robustness to new evidence (sample period extension) and new
economic analysis (hypotheses of explanatory variables)
Pagan Frontier
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
7 / 19
Structural interpretation and structural properties
Structural dimensions
Claims of structural representation of the macroeconomy can be
evaluated along several dimension, for example
I
I
I
I
Theoretical interpretation
Ability to explain data
Ability to explain earlier findings (properties of existing models)
Robustness to new evidence (sample period extension) and new
economic analysis (hypotheses of explanatory variables)
Pagan Frontier
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
7 / 19
Structural interpretation and structural properties
Structural dimensions
Claims of structural representation of the macroeconomy can be
evaluated along several dimension, for example
I
I
I
I
Theoretical interpretation
Ability to explain data
Ability to explain earlier findings (properties of existing models)
Robustness to new evidence (sample period extension) and new
economic analysis (hypotheses of explanatory variables)
Pagan Frontier
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
7 / 19
Structural interpretation and structural properties
Structural dimensions
Claims of structural representation of the macroeconomy can be
evaluated along several dimension, for example
I
I
I
I
Theoretical interpretation
Ability to explain data
Ability to explain earlier findings (properties of existing models)
Robustness to new evidence (sample period extension) and new
economic analysis (hypotheses of explanatory variables)
Pagan Frontier
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
7 / 19
Structural interpretation and structural properties
Structural dimensions
Claims of structural representation of the macroeconomy can be
evaluated along several dimension, for example
I
I
I
I
Theoretical interpretation
Ability to explain data
Ability to explain earlier findings (properties of existing models)
Robustness to new evidence (sample period extension) and new
economic analysis (hypotheses of explanatory variables)
Pagan Frontier
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
7 / 19
Structural interpretation and structural properties
A modified Pagan Frontier
’Index of structural properties’ replaces ’Empirical coherence’
along x-axis.
’Theoretical coherence’ not well defined in practice, since more
than one theoretical framework to consider (a least to the
heterodox).
I
I
Modern Walrasian theory (’microfoundations’, RBC, DSGE)
Post Walrasian, or ’Marshallian’ (stepwise modeling, heterogeniety,
involuntary unemployment, price makers, behavioural)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
8 / 19
Structural interpretation and structural properties
A modified Pagan Frontier
’Index of structural properties’ replaces ’Empirical coherence’
along x-axis.
’Theoretical coherence’ not well defined in practice, since more
than one theoretical framework to consider (a least to the
heterodox).
I
I
Modern Walrasian theory (’microfoundations’, RBC, DSGE)
Post Walrasian, or ’Marshallian’ (stepwise modeling, heterogeniety,
involuntary unemployment, price makers, behavioural)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
8 / 19
Structural interpretation and structural properties
A modified Pagan Frontier
’Index of structural properties’ replaces ’Empirical coherence’
along x-axis.
’Theoretical coherence’ not well defined in practice, since more
than one theoretical framework to consider (a least to the
heterodox).
I
I
Modern Walrasian theory (’microfoundations’, RBC, DSGE)
Post Walrasian, or ’Marshallian’ (stepwise modeling, heterogeniety,
involuntary unemployment, price makers, behavioural)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
8 / 19
Structural interpretation and structural properties
A modified Pagan Frontier
’Index of structural properties’ replaces ’Empirical coherence’
along x-axis.
’Theoretical coherence’ not well defined in practice, since more
than one theoretical framework to consider (a least to the
heterodox).
I
I
Modern Walrasian theory (’microfoundations’, RBC, DSGE)
Post Walrasian, or ’Marshallian’ (stepwise modeling, heterogeniety,
involuntary unemployment, price makers, behavioural)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
8 / 19
Structural interpretation and structural properties
A Modified Pagan Frontier
Degree of
theoretical
coherence
'Walras'
'Marshall'
Index of structural
properties
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
9 / 19
Evaluation of the New Keynesian Phillips Curve
The NKPC defined
e
∆pt = α∆pt+1
+ β∆pt−1 + δxt
(1)
Integral part of consensus monetary policy model. Represents the
supply side
The ultimate theory instigated inflation model (high on structural
interpretation
A declared success in terms of fit by proponents
...while the baseline pure forward looking model is rejected on
statistical grounds, it is still likely to be a reasonable first
approximation to inflation dynamics of both Europe and the
US, (Gali, 2003).
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
10 / 19
Evaluation of the New Keynesian Phillips Curve
The NKPC defined
e
∆pt = α∆pt+1
+ β∆pt−1 + δxt
(1)
Integral part of consensus monetary policy model. Represents the
supply side
The ultimate theory instigated inflation model (high on structural
interpretation
A declared success in terms of fit by proponents
...while the baseline pure forward looking model is rejected on
statistical grounds, it is still likely to be a reasonable first
approximation to inflation dynamics of both Europe and the
US, (Gali, 2003).
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
10 / 19
Evaluation of the New Keynesian Phillips Curve
The NKPC defined
e
∆pt = α∆pt+1
+ β∆pt−1 + δxt
(1)
Integral part of consensus monetary policy model. Represents the
supply side
The ultimate theory instigated inflation model (high on structural
interpretation
A declared success in terms of fit by proponents
...while the baseline pure forward looking model is rejected on
statistical grounds, it is still likely to be a reasonable first
approximation to inflation dynamics of both Europe and the
US, (Gali, 2003).
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
10 / 19
Evaluation of the New Keynesian Phillips Curve
The NKPC defined
e
∆pt = α∆pt+1
+ β∆pt−1 + δxt
(1)
Integral part of consensus monetary policy model. Represents the
supply side
The ultimate theory instigated inflation model (high on structural
interpretation
A declared success in terms of fit by proponents
...while the baseline pure forward looking model is rejected on
statistical grounds, it is still likely to be a reasonable first
approximation to inflation dynamics of both Europe and the
US, (Gali, 2003).
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
10 / 19
Evaluation of the New Keynesian Phillips Curve
Typical NKPC fits as well as a random-walk
Fitted values from NPC vs.
fitted values from random walk
Actual inflation and fit from the NPC
3.5
3.0
3.0
2.5
2.5
NPC-FIT
3.5
2.0
1.5
2.0
1.5
1.0
1.0
0.5
0.5
0.0
1975
1980
1985
NPCRATSFIT
1990
1995
0.0
0.0
0.5
1.0
DP
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
1.5
2.0
2.5
3.0
3.5
RW-FIT
EValW 2005
11 / 19
Evaluation of the New Keynesian Phillips Curve
Why is the NKPC only as good as a random-walk?
Low numerical significance of forcing variable (wage-share)
Hence dVAR feature of model dominates—also for hybrid
∆pt = 1.06∆pt−1 + 0.41∆2 pt−1 + 0.02wst−1
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
(2)
EValW 2005
12 / 19
Evaluation of the New Keynesian Phillips Curve
Why is the NKPC only as good as a random-walk?
Low numerical significance of forcing variable (wage-share)
Hence dVAR feature of model dominates—also for hybrid
∆pt = 1.06∆pt−1 + 0.41∆2 pt−1 + 0.02wst−1
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
(2)
EValW 2005
12 / 19
Evaluation of the New Keynesian Phillips Curve
Internal inconsistency of NKPC
Fuhrer (2005) analyzes acf of typical US NKPC
Inflation persistence is found to be due to ’intrinsic persistence’
not ’inherited persistence’ (from the forcing variable), as the theory
claims it should.
Hence, upon evaluation the theoretical content of the NKPC
’disappears out of the window’.
Belongs near zero in the modified Pagan Frontier (poor fit, no
structural interpretation after all)
How to model inflation is now in the blue (Fuhrer)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
13 / 19
Evaluation of the New Keynesian Phillips Curve
Internal inconsistency of NKPC
Fuhrer (2005) analyzes acf of typical US NKPC
Inflation persistence is found to be due to ’intrinsic persistence’
not ’inherited persistence’ (from the forcing variable), as the theory
claims it should.
Hence, upon evaluation the theoretical content of the NKPC
’disappears out of the window’.
Belongs near zero in the modified Pagan Frontier (poor fit, no
structural interpretation after all)
How to model inflation is now in the blue (Fuhrer)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
13 / 19
Evaluation of the New Keynesian Phillips Curve
Internal inconsistency of NKPC
Fuhrer (2005) analyzes acf of typical US NKPC
Inflation persistence is found to be due to ’intrinsic persistence’
not ’inherited persistence’ (from the forcing variable), as the theory
claims it should.
Hence, upon evaluation the theoretical content of the NKPC
’disappears out of the window’.
Belongs near zero in the modified Pagan Frontier (poor fit, no
structural interpretation after all)
How to model inflation is now in the blue (Fuhrer)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
13 / 19
Evaluation of the New Keynesian Phillips Curve
Internal inconsistency of NKPC
Fuhrer (2005) analyzes acf of typical US NKPC
Inflation persistence is found to be due to ’intrinsic persistence’
not ’inherited persistence’ (from the forcing variable), as the theory
claims it should.
Hence, upon evaluation the theoretical content of the NKPC
’disappears out of the window’.
Belongs near zero in the modified Pagan Frontier (poor fit, no
structural interpretation after all)
How to model inflation is now in the blue (Fuhrer)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
13 / 19
Evaluation of the New Keynesian Phillips Curve
Internal inconsistency of NKPC
Fuhrer (2005) analyzes acf of typical US NKPC
Inflation persistence is found to be due to ’intrinsic persistence’
not ’inherited persistence’ (from the forcing variable), as the theory
claims it should.
Hence, upon evaluation the theoretical content of the NKPC
’disappears out of the window’.
Belongs near zero in the modified Pagan Frontier (poor fit, no
structural interpretation after all)
How to model inflation is now in the blue (Fuhrer)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
13 / 19
Evaluation of the New Keynesian Phillips Curve
Encompassing
Wage bargaining models, (old) Phillips-curves and NKPC can be
(re)parameterized as EqC equations
NKPC:
e
∆pt = δ1f ∆pt+1
+ η1 ∆wst + δ1b ∆pt−1 − η1 [p − 1 (ws + p)]t−1 .
Bårdsen, Jansen and Nymoen (2004):
Encompass existing specifications with alternative EqCM part?
Encompass other with richer dynamics?
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
14 / 19
Evaluation of the New Keynesian Phillips Curve
Encompassing
Wage bargaining models, (old) Phillips-curves and NKPC can be
(re)parameterized as EqC equations
NKPC:
e
∆pt = δ1f ∆pt+1
+ η1 ∆wst + δ1b ∆pt−1 − η1 [p − 1 (ws + p)]t−1 .
Bårdsen, Jansen and Nymoen (2004):
Encompass existing specifications with alternative EqCM part?
Encompass other with richer dynamics?
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
14 / 19
Evaluation of the New Keynesian Phillips Curve
Encompassing
Wage bargaining models, (old) Phillips-curves and NKPC can be
(re)parameterized as EqC equations
NKPC:
e
∆pt = δ1f ∆pt+1
+ η1 ∆wst + δ1b ∆pt−1 − η1 [p − 1 (ws + p)]t−1 .
Bårdsen, Jansen and Nymoen (2004):
Encompass existing specifications with alternative EqCM part?
Encompass other with richer dynamics?
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
14 / 19
Evaluation of the New Keynesian Phillips Curve
Encompassing
Wage bargaining models, (old) Phillips-curves and NKPC can be
(re)parameterized as EqC equations
NKPC:
e
∆pt = δ1f ∆pt+1
+ η1 ∆wst + δ1b ∆pt−1 − η1 [p − 1 (ws + p)]t−1 .
Bårdsen, Jansen and Nymoen (2004):
Encompass existing specifications with alternative EqCM part?
Encompass other with richer dynamics?
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
14 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Alternative (non Walrasias) models of wage and price setting
jointly have been around for a long time
I
I
I
Norwegian model of inflation
Phillips curve
Game theory, monopolistic price setting
Starting from a premise that wages, prices and productivity are
I(1), and rate of unemployment I(0), but subject to deterministic
shifts, possible to encompass the different models in a identified
cointegration model.
I
I
CI and error correction singularly ’in’ unemployment gives Phillips
curve
CI and error correction in wages (and prices) gives fair-game
outcome a la Hahn and Solow (1997). Equilibrium real wage locus
is ’thick’, consistent with different unemployment rates
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
15 / 19
Evaluation of an econometric model of the inflation spiral
Example of identified steady-state
Data for log of Norwegian CPI (p), wage rate (w), productivity (z),
import price (pb) and unemployment rate (u). 1972(4)-2004(2).
Applying the structural interpretation of the theory to the
cointegration results, gives the following identified steady-state
relationships (simplifid from Table 2, p 30)
w − p − z = µw − 0.15u
(3)
w − p − z = µp − 0.4(w − z − pb)
(4)
Tempting but invalid to solve for NAIRU from this system.
Instead, specify error-correction model of inflation spiral using (3)
and (4) and general-to-specific modelling of I(0), system.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
16 / 19
Evaluation of an econometric model of the inflation spiral
Example of identified steady-state
Data for log of Norwegian CPI (p), wage rate (w), productivity (z),
import price (pb) and unemployment rate (u). 1972(4)-2004(2).
Applying the structural interpretation of the theory to the
cointegration results, gives the following identified steady-state
relationships (simplifid from Table 2, p 30)
w − p − z = µw − 0.15u
(3)
w − p − z = µp − 0.4(w − z − pb)
(4)
Tempting but invalid to solve for NAIRU from this system.
Instead, specify error-correction model of inflation spiral using (3)
and (4) and general-to-specific modelling of I(0), system.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
16 / 19
Evaluation of an econometric model of the inflation spiral
Example of identified steady-state
Data for log of Norwegian CPI (p), wage rate (w), productivity (z),
import price (pb) and unemployment rate (u). 1972(4)-2004(2).
Applying the structural interpretation of the theory to the
cointegration results, gives the following identified steady-state
relationships (simplifid from Table 2, p 30)
w − p − z = µw − 0.15u
(3)
w − p − z = µp − 0.4(w − z − pb)
(4)
Tempting but invalid to solve for NAIRU from this system.
Instead, specify error-correction model of inflation spiral using (3)
and (4) and general-to-specific modelling of I(0), system.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
16 / 19
Evaluation of an econometric model of the inflation spiral
Example of identified steady-state
Data for log of Norwegian CPI (p), wage rate (w), productivity (z),
import price (pb) and unemployment rate (u). 1972(4)-2004(2).
Applying the structural interpretation of the theory to the
cointegration results, gives the following identified steady-state
relationships (simplifid from Table 2, p 30)
w − p − z = µw − 0.15u
(3)
w − p − z = µp − 0.4(w − z − pb)
(4)
Tempting but invalid to solve for NAIRU from this system.
Instead, specify error-correction model of inflation spiral using (3)
and (4) and general-to-specific modelling of I(0), system.
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
16 / 19
Evaluation of an econometric model of the inflation spiral
Final equilibrium correction model, p. 30

1 + 0.26L

0
(0.06)




− 0.21 + 0.18 L
1 − 0.11 L2
(0.03)

0
(0.03)
0.09
(0.05)
0

"
d
∆w
c
∆p
#
=
t

∆2 y
  ∆z 
0.03L − 0.03 0.025
∆pb t
(0.01)
(0.006) (0.008)


0.12
0
(0.01)
 ecmw
−
0
0.06
ecmp t−1
(0.015)
(5)

(0.008)
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
17 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Final model: some structural properties
Integrates identified steady-state relationships
Valid simplification of unrestricted I(0) VAR
The two equation are ’recognizable’ as wage-growth and price
adjustment equations, e.g.,∆wt reacts to ecmw,t−1 and vice versa.
Represents ’changes in centralization’ of wage-setting
Derivative coefficients of model constant across regimes
Identified final ecm has a recursive ’structure’.
Phillips-curve interpretation statistically rejected (confirming
cointegration)
Not encompassed by NKPC
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
18 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Evaluation of an econometric model of the inflation spiral
Stepwise modelling
Final model is (of course) an incomplete model of the inflation
spiral.
Hence a structural representation of the inflation spiral must be
achieve by stepwise modelling.
I
I
I
I
In line with traditions of macroeconometric model building
Opposite methodology of the one-step modelling in Walrasian
macroeconomics
Relationship to Copenhagen methodology? Only a matter of size?
Intuitive
If a model of the whole economy is to be securely based,
it must be grounded in an intelligible account of how a single
market is supposed to work, Hicks (1965), p. 78
Ragnar Nymoen ( http://folk.uio.no/rnymoen/
AssessingDepartment
structure in monetary
of Economics
policyUniversity
models of Oslo )
EValW 2005
19 / 19
Download