USING AUTOMETRICS

Genaro Sucarrat
(Department of Economics, UC3M)

http://www.eco.uc3m.es/sucarrat/

March 25, 2008

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• The basics of OxMetrics 5:

→ Loading, editing and transforming data, creating “special”

series (cointegration relations, trends, etc.)

• An overview of Autometrics:

→ Key concepts and characteristics

• Single-equation modelling with Autometrics

→ Formulation, Advanced Autometrics settings, fixing variables,

example (2007 Econometric Game, Q1)

• Multiple-equation modelling with Autometrics

→ Formulation, fixing variables, example (2007 Econometric

Game, Q2)
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

OxMetrics basics:

• Load data: File → Open, etc.

• Edit sample/dates: Edit → Change Sample

• Missing values (recommendation): Set to “missing” by

double-clicking the data cell in question

• Graph series: Model → Graphics (or click on the graphics

button) → Actual series or All plot types

• Transform data (algebra feature):

Edit → Algebra (or click on the “Alg” button) →

“Code”, e.g. “DCOO = diff(COO,1);” → Run (→ Done)

(NOTE: Case sensitivity in variable names!)

• Create special series (calculator feature): Model → Calculator
(or click on the calculator button) → . . .
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Example. Edit dates (2007 Econometric Game Case):

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Example. Create a differenced series:

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

An overview of Autometrics:

• The objective of Autometrics is to automate General-to-Specific

(Gets) multiple path simplification search of a data coherent,
General Unrestricted Model (GUM) in the form of a linear OLS/IV
estimable regression (or regressions)

• Default definition of data-coherency: Stable parameters and

Gaussian, serially uncorrelated, homoscedastic errors. NOTE:
These assumptions can be relaxed through the “Advanced
Autometrics settings”, and if the GUM fails one or several
diagnostic checks Autometrics proceeds anyway

• GUM: A general model (advice: Not too general!) that includes

the variables and lags that are believed to possibly have an impact

• Further reading: Doornik and Hendry (2007a, pp. 70-77),

Hendry and Krolzig (2001) (Autometrics is an evolution of PcGets)
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Key benefits of Gets modelling:

→ Estimation and inference while controlling for the impact from
other variables: Spurious variables are more likely to be excluded,
parameter estimates are “more correct”
→ Gets modelling leads to a parsimonious, explanatory model
particularly useful for scenario analysis (policy analysis, counter-
factual analysis and conditional forecasting)
• Originally, the main disadvantage of Gets modelling was:
→ Resource demanding and time consuming if properly imple-
mented
• Solution: Automated Gets
→ Hoover and Perez (1999), PcGets (Hendry and Krolzig 2001),
Autometrics (Doornik 2007, Doornik and Hendry 2007a)
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

What does multiple path Gets specification search consist in?

• Sequential deletion of insignificant variables (significance level

can be user-specified), while checking data-coherency for each
variable deletion (variable-deletion that induces data-incoherency,
say, serially correlated errors, is not undertaken)

• Multiple path Gets search typically leads to several terminal

models; Autometrics either applies Gets on their union or chooses
among them by means of information criteria (can be
user-specified; keyword: “Tie-breaker”)
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

Single-equation modelling with Autometrics. Example: 2007

Econometric Game, Question 1
• My GUM:
11
X
∆COOt = b0 + b1 ∆COOt−1 + b2 ∆COOt−2 + cj dj,t + et
j=1

• The specific model proposed by Autometrics using the default

options:
3
X 11
X
∆COOt = b0 + b1 ∆COOt−1 + cj dj,t + cj dj,t + et
j=1 j=5

• Unfortunately, the missing values cannot be estimated by means

of the level representation
3
X 11
X
COOt = b0 +(1+b1 )COOt−1 +b1 COOt−2 + cj dj,t + cj dj,t +et
j=1 j=5

so a different specification is needed with, say, no lags of ∆COOt

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Formulate a model: (Model →) PcGive → Category: “Models

for time series data” → Model class: “Single-equation dynamic
modelling using PcGive” → “Formulate”
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Specify model:

USEFUL FEATURE: Fixing regressors (that is, preventing

Autometrics from deleting them). Select the regressors to fix →
Right-click mouse → A: instrument/fixed. NOTE: This defines
instrument if IV is used instead of OLS
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Selected Autometrics options:

→ Target size: Significance level

→ Outlier detection: Neutralises large residuals in the GUM by

means of impulse dummies

→ Pre-search lag reduction: Speeds up simplification; GENERAL

ADVICE: Turn off!
→ Advanced Autometrics settings: In general, use only if default
settings and options are unsatisfactory

→ Recursive graphics: TURN ON! Slows down computation

(slightly), but enables some very useful stability diagnostics
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Advanced Autometrics settings:

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

• Specific model proposed by Autometrics:

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

Further diagnostic tests:

• Residuals graphs: Model → Test → Graphical analysis → . . .

• User specified residuals tests: Model → Test → Test → . . .

• Recursive graphics (VERY useful!): Model → Test → Recursive

graphics → . . .
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

Multiple-equation modelling with Autometrics: Two possibilities

• Seemingly Unrelated Regression (SUR) using OLS/IV, that is,

single-equation modelling of each equation separately (requires
stationarity of regressors)

• Simultaneous variable deletion (or non-deletion) across equations

using vector diagnostic tests but estimation still by OLS (does not
require stationarity of regressors), see Doornik and Hendry (2007b,
pp. 29-31). (NOTE: IV estimation not available with this strategy)

→ Model type: “Unrestricted system” (system of URFs), see

Doornik and Hendry (2007b, chapter 3)
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

Formulate a system: (Model →) PcGive → Category: “Models for

time series data” → Model class: “Multiple-equation dynamic
modelling using PcGive” → “Formulate”
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

Multiple-equation modelling with Autometrics using second

possibility. Example: 2007 Econometric Game, Question 2

• My GUM: A six-dimensional VAR(2) of

yt = (TEMPt , COOt , RADt , PREt , VAPt , CLDt ), with a constant
and 11 centered seasonals in each of the six equations:

NOTE: Fixing variables (that is, not allowing Autometrics to delete

them) now differs. Select the variables to delete → Right-click
mouse → U: Unrestricted. (Unfixing: Z: regressor)
Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

Results:

• NOTE: Autometrics simplifies even though the GUM does not

pass all diagnostic checks

• Four variables are removed from all of the equations: The second
lag of TEMP, VAP and CLD, and CSeasonal 10

Other type of analysis:

• Cointegration analysis (applied on the Unrestricted system, not

on the simplified model): Model → Test → Dynamic Analysis and
Cointegration Tests → . . .

See Doornik and Hendry (2007b, chapter 4)

Introduction OxMetrics basics Autometrics Single-equation modelling Multiple-equation modelling References

References:
Doornik, J. (2007). Autometrics. Working Paper, Economics
Department, University of Oxford.
Doornik, J. A. and D. F. Hendry (2007a). Empirical Econometric
Modelling - PcGive 12: Volume I. London: Timberlake
Consultants Ltd.
Doornik, J. A. and D. F. Hendry (2007b). Empirical Econometric
Modelling - PcGive 12: Volume II. London: Timberlake
Consultants Ltd.
Hendry, D. F. and H.-M. Krolzig (2001). Automatic Econometric
Model Selection using PcGets. London: Timberlake Consultants
Press.
Hoover, K. D. and S. J. Perez (1999). Data Mining Reconsidered:
Encompassing and the General-to-Specific Approach to
Specification Search. Econometrics Journal 2, 167–191.