Atjaunināt sīkdatņu piekrišanu

Time Series Econometrics: using Microfit 5.0 [Mīkstie vāki]

(Research Consultant at Wadhwani Asset Management), (Professor of Economics, University of Cambridge)
  • Formāts: Paperback / softback, 592 pages, height x width x depth: 244x188x31 mm, weight: 1112 g, numerous figures and tables
  • Izdošanas datums: 27-Aug-2009
  • Izdevniecība: Oxford University Press
  • ISBN-10: 0199563535
  • ISBN-13: 9780199563531
Citas grāmatas par šo tēmu:
  • Mīkstie vāki
  • Cena: 72,63 €*
  • * Šī grāmata vairs netiek publicēta. Jums tiks paziņota lietotas grāmatas cena
  • Šī grāmata vairs netiek publicēta. Jums tiks paziņota lietotas grāmatas cena.
  • Daudzums:
  • Ielikt grozā
  • Pievienot vēlmju sarakstam
Time Series Econometrics: using Microfit 5.0Palielināt
  • Formāts: Paperback / softback, 592 pages, height x width x depth: 244x188x31 mm, weight: 1112 g, numerous figures and tables
  • Izdošanas datums: 27-Aug-2009
  • Izdevniecība: Oxford University Press
  • ISBN-10: 0199563535
  • ISBN-13: 9780199563531
Citas grāmatas par šo tēmu:
Permanent link: https://www.kriso.lv/db/9780199563531.html
Keywords:
Microfit 5.0 is an interactive, menu-driven program with a host of facilities for estimation, hypothesis testing, forecasting, data processing, file management, and graphic display. These features make Microfit 5.0 one of the most powerful menu-driven time-series econometric packages currently available. It is a major advance over Microfit 4 and offers a unique built-in interactive, searchable econometric text. It provides users with technical, functional and tutorial help throughout the package. Another key feature is that it can be used at different levels of technical sophistication. For experienced users of econometric programs, it offers a variety of univariate methods, multivariate techniques for cointegration, principal components, canonical correlations and multivariate volatility modelling, and provides a large number of diagnostic and non-nested tests not readily available in other packages. The interaction of excellent graphics and estimation capabilities in Microfit 5.0 allows important econometric research to be carried out in a matter of days rather than weeks.

This comprehensive and accessible manual outlines and explains all of Microfit's features and functionality and is a valuable resource in its own right for those new to Microfit and those who want to use and understand its more advanced features.

Microfit 5.0 is an interactive, menu-driven program with a host of facilities for estimation, hypothesis testing, forecasting, data processing, file management, and graphic display. These features make Microfit 5.0 one of the most powerful menu-driven time-series econometric packages currently available. It is a major advance over Microfit 4 and offers a unique built-in interactive, searchable econometric text. It provides users with technical, functional and tutorial help throughout the package. Another key feature is that it can be used at different levels of technical sophistication. For experienced users of econometric programmes, it offers a variety of univariate methods, multivariate techniques for cointegration, principal components, canonical correlations and multivariate volatility modelling, and provides a large number of diagnostic and non-nested tests not readily available in other packages. The interaction of excellent graphics and estimation capabilities in Microfit 5.0 allows important econometric research to be carried out in a matter of days rather than weeks.

This comprehensive and accessible manual outlines and explains all of Microfit's features and functionality and is a valuable resource in its own right for those new to Microfit and those who want to use and understand its more advanced features.
I Introduction to Microfit
1(20)
1 Introduction
3(10)
1.1 What is Microfit?
3(1)
1.2 New Features of Microfit 5.0
3(4)
1.2.1 New functions and commands
5(1)
1.2.2 Single equation estimation techniques
5(1)
1.2.3 System equation estimation techniques
6(1)
1.3 Tutorial lessons
7(1)
1.4 Other features of Microfit 5.0
8(4)
1.4.1 Data management
8(1)
1.4.2 Data transformations
8(1)
1.4.3 High-resolution graphics
8(1)
1.4.4 Batch operations
8(1)
1.4.5 General statistics
9(1)
1.4.6 Dynamic simulation
9(1)
1.4.7 Other single equation estimation techniques
9(1)
1.4.8 Model respecification
10(1)
1.4.9 Diagnostic tests and model selection criteria
10(1)
1.4.10 Variable addition and variable deletion tests
11(1)
1.4.11 Cointegration tests
11(1)
1.4.12 Testing for unit roots
11(1)
1.4.13 Tests of linear and non-linear restrictions
11(1)
1.4.14 Non-nested tests
11(1)
1.4.15 Static and dynamic univariate forecasts
11(1)
1.5 Installation and system configuration
12(1)
1.6 System requirements for Microfit 5.0
12(1)
2 Installation and Getting Started
13(8)
2.1 Installation
13(1)
2.1.1 Single user installation
13(1)
2.1.2 Network installation
13(1)
2.2 Starting and ending a session
14(1)
2.2.1 Running Microfit
14(1)
2.2.2 Quitting Microfit
14(1)
2.3 Using windows, menus and buttons
14(3)
2.3.1 The main window
14(1)
2.3.2 Main Menu bar
14(2)
2.3.3 Buttons
16(1)
2.4 The Variables window
17(1)
2.5 The Data window
18(3)
2.5.1 Program options
18(1)
2.5.2 Help
19(2)
II Processing and Data Management
21(52)
3 Inputting and Saving Data Files
23(14)
3.1 Change data dimension
23(1)
3.2 Inputting data
23(7)
3.2.1 Inputting data from the keyboard
24(2)
3.2.2 Loading an existing data set
26(1)
3.2.3 Inputting data from a raw data (ASCII) file
26(1)
3.2.4 Inputting data from a special Microfit file saved previously
27(1)
3.2.5 Inputting data from an Excel file
27(1)
3.2.6 Inputting data from CSV files
28(1)
3.2.7 Inputting data from AREMOS (TSD) files
28(1)
3.2.8 Input new data from the clipboard into Microfit
28(1)
3.2.9 Adding data from the clipboard into Microfit workspace
29(1)
3.2.10 Inputting daily data
30(1)
3.3 Adding two data files
30(4)
3.3.1 Adding two special Microfit files containing the same variables
31(2)
3.3.2 Adding two special Microfit files containing different variables
33(1)
3.4 Using the Commands and Data Transformations box
34(1)
3.5 Saving data
34(2)
3.5.1 Save as a special Microfit file
34(1)
3.5.2 Save as an Excel sheet
35(1)
3.5.3 Save as a comma separated values (CSV) file
35(1)
3.5.4 Save as an AREMOS (TSD) file
36(1)
3.5.5 Save as a raw data (numbers only) file
36(1)
3.6 Starting with a new data set
36(1)
4 Data Processing and Preliminary Data Analysis
37(31)
4.1 Creating constant terms, time trends and seasonal dummies
38(3)
4.1.1 Creating a constant (intercept) term
39(1)
4.1.2 Creating a time trend
39(1)
4.1.3 Creating (0, 1) seasonal dummies
39(1)
4.1.4 Creating centred seasonal dummies
40(1)
4.1.5 Creating seasonal dummies relative to the last season
40(1)
4.2 Typing formulae in Microfit
41(2)
4.2.1 Printing, saving, viewing, and copying files
43(1)
4.3 Using built-in functions in Microfit
43(9)
4.3.1 Function ABS
43(1)
4.3.2 Function COS
43(1)
4.3.3 Function CPHI
44(1)
4.3.4 Function CSUM
44(1)
4.3.5 Function EXP
44(1)
4.3.6 Function GDL
44(1)
4.3.7 Function HPF
45(1)
4.3.8 Function INVNORM
45(1)
4.3.9 Function LOG
46(1)
4.3.10 Function MAX
46(1)
4.3.11 Function MAV
46(1)
4.3.12 Function MEAN
46(1)
4.3.13 Function MIN
47(1)
4.3.14 Function NORMAL
47(1)
4.3.15 Function ORDER
47(1)
4.3.16 Function PHI
47(1)
4.3.17 Function PTTEST
48(1)
4.3.18 Function RANK
48(1)
4.3.19 Function RATE
48(1)
4.3.20 Function REC_MAX
49(1)
4.3.21 Function REC_MIN
49(1)
4.3.22 Function ROLL_MAX
49(1)
4.3.23 Function ROLL_MIN
49(1)
4.3.24 Function SIGN
50(1)
4.3.25 Funnction SIN
50(1)
4.3.26 Function SORT
50(1)
4.3.27 Function SQRT
51(1)
4.3.28 Function STD
51(1)
4.3.29 Function SUM
51(1)
4.3.30 Function UNIFORM
51(1)
4.4 Using commands in Microfit
52(16)
4.4.1 Command ADD
52(1)
4.4.2 Command ADF
52(2)
4.4.3 Command ADF_GLS
54(1)
4.4.4 Command ADF_MAX
55(1)
4.4.5 Command ADF_WS
55(1)
4.4.6 Command BATCH
56(1)
4.4.7 Command CCA
57(1)
4.4.8 Command COR
58(1)
4.4.9 Command DELETE
58(1)
4.4.10 Command DF_PP
58(1)
4.4.11 Command ENTITLE
59(1)
4.4.12 Command FILL_FORWARD
59(1)
4.4.13 Command FILL_MISSING
60(1)
4.4.14 Command HIST
60(1)
4.4.15 Command KEEP
60(1)
4.4.16 Command KPSS
61(1)
4.4.17 Command LIST
61(1)
4.4.18 Command NONPARM
61(2)
4.4.19 Command PCA
63(1)
4.4.20 Command PLOT
63(1)
4.4.21 Command REORDER
64(1)
4.4.22 Command RESTORE
64(1)
4.4.23 Command SAMPLE
64(1)
4.4.24 Command SCATTER
65(1)
4.4.25 Command SIM
65(1)
4.4.26 Command SIMB
66(1)
4.4.27 Command SPECTRUM
66(1)
4.4.28 Command TITLE
67(1)
4.4.29 Command XPLOT
67(1)
5 Printing/Saving Results and Graphs
68(5)
5.1 Result screens
68(1)
5.1.1 On-line printing of results
68(1)
5.1.2 Saving results
68(1)
5.2 Print/save/retrieve graphs
69(3)
5.2.1 Altering the display of graphs
70(1)
5.2.2 Printing graphs
71(1)
5.2.3 Saving graphs
71(1)
5.2.4 Retrieval of graphic files
72(1)
5.2.5 Capturing graphs onto the clipboard
72(1)
5.3 Exercises using graphs
72(1)
5.3.1 Exercise 5.1
72(1)
5.3.2 Exercise 5.2
72(1)
III Estimation Menus
73(84)
6 Single-Equation Options
75(42)
6.1 The classical normal linear regression model
75(4)
6.1.1 Testing the assumptions of the classical model
76(1)
6.1.2 Estimation of the classical linear regression model
77(2)
6.1.3 Testing zero restrictions and reporting probability values
79(1)
6.2 The maximum likelihood approach
79(4)
6.2.1 Newton-Raphson algorithm
80(1)
6.2.2 Properties of maximum likelihood estimators
80(1)
6.2.3 Likelihood-based tests
81(2)
6.3 Estimation menus in Microfit
83(1)
6.4 Single Equation Estimation Menu
84(1)
6.5 The Linear Regression Menu
85(3)
6.5.1 Specification of a linear regression equation
85(2)
6.5.2 Specification of the estimation period
87(1)
6.6 Ordinary Least Squares option
88(2)
6.6.1 Tests of residual serial correlation
88(1)
6.6.2 Ramsey's RESET test for functional form misspecification
89(1)
6.6.3 The normality test
89(1)
6.6.4 Heteroscedasticity test
89(1)
6.6.5 Predictive failure test
89(1)
6.6.6 Chow's test of the stability of regression coefficients
90(1)
6.6.7 Measures of leverage
90(1)
6.7 Generalized instrumental variable method option
90(2)
6.8 AR errors (exact ML) option
92(1)
6.9 AR errors (Cochrane-Orcutt) option
93(1)
6.10 AR errors (Gauss-Newton) option
94(1)
6.11 IV with AR errors (Gauss-Newton) option
94(2)
6.12 MA errors (exact ML) option
96(1)
6.13 IV with MA errors option
96(1)
6.13.1 Specification of initial estimates for the parameters of the AR/MA error process
97(1)
6.14 Recursive regression options
97(1)
6.14.1 Recursive OLS Regression Results Menu
98(1)
6.15 Rolling Linear Regression Menu
98(1)
6.15.1 Rolling Regression Results Menu
99(1)
6.16 Non-Linear Regression Menu
99(3)
6.16.1 Specification of a non-linear regression equation
101(1)
6.16.2 Specification of initial parameter estimates
102(1)
6.16.3 Estimation results for the non-linear regression equation
102(1)
6.17 Phillips-Hansen Estimation Menu
102(2)
6.18 ARDL approach to cointegration
104(2)
6.18.1 Specification of an ARDL regression equation
104(1)
6.18.2 ARDL Order Selection Menu
105(1)
6.18.3 Post ARDL Model Selection Menu
105(1)
6.18.4 ARDL Forecast Menu
106(1)
6.19 Logit and Probit models
106(3)
6.19.1 Specification of the Logit/Probit model
107(1)
6.19.2 Logit/Probit Estimation Menu
107(1)
6.19.3 Estimation results for Logit and Probit options
108(1)
6.19.4 Logit/Probit Post Estimation Menu
108(1)
6.20 Post Regression Menu
109(1)
6.21 Display/Save Residuals and Fitted Values Menu
110(1)
6.22 Standard, White and Newey-West Adjusted Variance Menu
111(1)
6.23 Hypothesis Testing Menu
112(5)
7 Multiple Equation Options
117(26)
7.1 The canonical multivariate model
117(2)
7.1.1 The log-likelihood function of the multivariate model
119(1)
7.2 General guidelines
119(2)
7.3 System Estimation Menu
121(1)
7.4 Unrestricted VAR option
122(6)
7.4.1 Unrestricted VAR Post Estimation Menu
123(1)
7.4.2 Unrestricted VAR Dynamic Response Analysis Menu
124(1)
7.4.3 VAR Hypothesis Testing Menu
125(2)
7.4.4 Multivariate Forecast Menu
127(1)
7.5 Cointegrating VAR options
128(8)
7.5.1 Specification of the cointegrating VAR model
130(1)
7.5.2 Cointegrating VAR Post Estimation Menu
131(1)
7.5.3 Long-Run Structural Modelling Menu
132(1)
7.5.4 Impulse Response Analysis and Forecasting Menu
133(3)
7.5.5 Beveridge-Nelson Trend/Cycle Decomposition
136(1)
7.5.6 Trend/Cycle Decomposition Results Menu
136(1)
7.6 Cointegrating VARX option
136(2)
7.7 SURE options
138(5)
7.7.1 Unrestricted SURE options
139(1)
7.7.2 Restricted SURE options
140(1)
7.7.3 SURE Post Estimation Menu
141(2)
8 Volatility Modelling Options
143(14)
8.1 Introduction
143(1)
8.2 Historical approaches to volatility measurement
144(1)
8.2.1 Risk Metrics TM (JP Morgan) method
144(1)
8.2.2 Econometric approaches
145(1)
8.3 Univariate GARCH models
145(1)
8.4 Multivariate GARCH models
146(2)
8.4.1 DCC and t-DCC Multivariate Volatility Models
147(1)
8.5 Volatility Modelling Menu
148(1)
8.6 Univariate GARCH Estimation Menu
149(2)
8.6.1 Specification of the GARCH, AGARCH and EGARCH models
149(1)
8.6.2 Specification of the initial parameter values for GARCH, AGARCH and EGARCH models
150(1)
8.6.3 Estimation results for the GARCH-M options
150(1)
8.7 Multivariate GARCH Menu
151(2)
8.7.1 Estimation results for the MGARCH
152(1)
8.8 Multivariate GARCH Post Estimation Menu
153(4)
8.8.1 Testing the Validity of Multivariate GARCH Menu
154(1)
8.8.2 Compute the VaR of a portfolio
154(3)
IV Tutorial Lessons
157(260)
9 Lessons in Data Management
159(7)
9.1 Lesson 9.1: Reading in the raw data file UKSTOCK.DAT
159(1)
9.2 Lesson 9.2: Saving your current data set as a special Microfit file
160(1)
9.3 Lesson 9.3: Reading in the special Microfit file UKSTOCK.FIT
160(1)
9.4 Lesson 9.4: Combining two special Microfit files containing different variables
161(1)
9.5 Lesson 9.5: Combining two special Microfit files containing the same variables
161(1)
9.6 Lesson 9.6: Extending the sample period of a special Microfit file
162(1)
9.7 Lesson 9.7: Reading the CSV file UKCON.CSV into Microfit
162(1)
9.8 Lesson 9.8: Reading the Excel file DAILYFUTURES.XLS into Microfit
163(1)
9.9 Lesson 9.9: Saving the DAILYFUTURES.XLS file excluding missing values
163(1)
9.10 Exercises in data management
164(2)
9.10.1 Exercise 9.1
164(1)
9.10.2 Exercise 9.2
164(1)
9.10.3 Exercise 9.3
164(1)
9.10.4 Exercise 9.4
165(1)
10 Lessons in Data Processing
166(31)
10.1 Lessons 10.1: Interactive data transformations
166(1)
10.2 Lessons 10.2: Doing data transformations using the BATCH command
167(1)
10.3 Lessons 10.3: Adding titles (descriptions) to variables
168(1)
10.4 Lessons 10.4: Creating dummy variables
169(2)
10.5 Lessons 10.5: Plotting variables against time and/or against each other
171(2)
10.6 Lessons 10.6: The use of command XPLOT in generating probability density function
173(1)
10.7 Lessons 10.7: Histogram of US stock market returns
174(1)
10.8 Lessons 10.8: Hodrick-Prescott filter applied to UK GDP
175(2)
10.9 Lessons 10.9: Summary statistics and correlation coefficients of US and UK output growths
177(1)
10.10 Lessons 10.10: Autocorrelation coefficients of US output growth
178(1)
10.11 Lessons 10.11: Spectral density function of the US output growth
179(2)
10.12 Lessons 10.12: Constructing a geometrically declining distributed lag variable: using the SIM command
181(1)
10.13 Lesson 10.13: Computation of OLS estimators using formulae and commands
182(3)
10.14 Lessons 10.14: Construction of indices of effective exchange rates and foreign prices
185(3)
10.15 Lessons 10.15: Non-parametric density estimation of futures returns
188(3)
10.16 Lessons 10.16: Principal components analysis of US macro-economic time series
191(3)
10.17 Lesson 10.17: Canonical correlation analysis of bond and equity futures
194(2)
10.18 Exercises in data processing
196(1)
10.18.1 Exercise 10.1
196(1)
10.18.2 Exercise 10.2
196(1)
10.18.3 Exercise 10.3
196(1)
10.18.4 Exercise 10.4
196(1)
11 Lessons in Linear Regression Analysis
197(35)
11.1 Lesson 11.1: OLS estimation of simple regression models
197(5)
11.2 Lesson 11.2: Two alternative methods of testing linear restrictions
202(4)
11.3 Lesson 11.3: Estimation of long-run effects and mean lags
206(2)
11.4 Lesson 11.4: The multicollinearity problem
208(4)
11.5 Lesson 11.5: Testing common factor restrictions
212(1)
11.6 Lesson 11.6: Estimation of regression models with serially correlated errors
213(4)
11.7 Lesson 11.7: Estimation of a `surprise' consumption function: an example of two-step estimation
217(2)
11.8 Lesson 11.8: An example of non-nested hypothesis testing
219(1)
11.9 Lesson 11.9: Testing linear versus log-linear models
220(2)
11.10 Lesson 11.10: Testing for exogeneity: computation of the Wu-Hausman statistic
222(2)
11.11 Lesson 11.11: Recursive prediction of US monthly excess returns
224(4)
11.12 Lesson 11.12: Rolling regressions and the Lucas critique
228(2)
11.13 Exercises in linear regression analysis
230(2)
11.13.1 Exercise 11.1
230(1)
11.13.2 Exercise 11.2
231(1)
11.13.3 Exercise 11.3
231(1)
11.13.4 Exercise 11.4
231(1)
11.13.5 Exercise 11.5
231(1)
11.13.6 Exercise 11.6
231(1)
12 Lessons in Univariate Time-Series Analysis
232(23)
12.1 Lesson 12.1: Using the ADF command to test for unit roots
233(4)
12.2 Lesson 12.2: Spectral analysis of US output growth
237(6)
12.3 Lesson 12.3: Using an ARMA model for forecasting US output growth
243(3)
12.4 Lesson 12.4: Alternative measures of persistence of shocks to US real GNP
246(3)
12.5 Lesson 12.5: Non-stationarity and structural breaks in real GDP
249(2)
12.6 Lesson 12.6: Unit roots in US nominal wages and the stock market crash
251(3)
12.7 Exercises in univariate time-series analysis
254(1)
12.7.1 Exercise 12.1
254(1)
12.7.2 Exercise 12.2
254(1)
12.7.3 Exercise 12.3
254(1)
13 Lessons in Non-Linear Estimation
255(15)
13.1 Lesson 13.1: Non-linear estimation of Cobb-Douglas production function
255(2)
13.2 Lesson 13.2: Estimation of Euler equations by the NLS-IV method
257(4)
13.3 Lesson 13.3: Estimation of Almon distributed lag models
261(2)
13.4 Lesson 13.4: Estimation of a non-linear Phillips curve
263(4)
13.5 Lesson 13.5: Estimating a non-linear Phillips curve with serially correlated errors
267(2)
13.6 Exercises in non-linear estimation
269(1)
13.6.1 Exercise 13.1
269(1)
13.6.2 Exercise 13.2
269(1)
13.6.3 Exercise 13.3
269(1)
13.6.4 Exercise 13.4
269(1)
14 Lessons in Probit and Logit Estimation
270(7)
14.1 Lesson 14.1: Modeling the choice of fertilizer use by Philippine farmers
270(4)
14.1.1 Forecasting with Probit/Logit models
273(1)
14.2 Lesson 14.2: Fertilizer use model estimated over a sub-sample of farmers
274(2)
14.3 Exercises in Logit/Probit estimation
276(1)
14.3.1 Exercise 14.1
276(1)
14.3.2 Exercise 14.2
276(1)
15 Lessons in VAR Modelling
277(13)
15.1 Lesson 15.1: Selecting the order of the VAR
277(4)
15.2 Lesson 15.2: Testing for the presence of oil shock dummies in output equations
281(1)
15.3 Lesson 15.3: International transmission of output shocks
282(1)
15.4 Lesson 15.4: Contemporaneous correlation of output shocks
283(2)
15.5 Lesson 15.5: Forecasting output growths using the VAR
285(1)
15.6 Lesson 15.6: Impulse responses of the effects of output growth shocks
286(2)
15.7 Exercises in VAR modelling
288(2)
15.7.1 Exercise 15.1
288(1)
15.7.2 Exercise 15.2
288(1)
15.7.3 Exercise 15.3
289(1)
16 Lessons in Cointegration Analysis
290(49)
16.1 Lesson 16.1: Testing for cointegration when the cointegrating coefficients are known
291(3)
16.2 Lesson 16.2: A residual-based approach to testing for cointegration
294(2)
16.3 Lesson 16.3: Testing for cointegration: Johansen ML approach
296(6)
16.4 Lesson 16.4: Testing for cointegration in models with I(1) exogenous variables
302(6)
16.5 Lesson 16.5: Long-run analysis of consumption, income and inflation: the ARDL approach
308(4)
16.6 Lesson 16.6: Great ratios and long-run money demand in the US
312(14)
16.7 Lesson 16.7: Application of the cointegrating VAR analysis to the UK term structure of interest rates
326(8)
16.8 Lesson 16.8: Canonical correlations and cointegration analysis
334(3)
16.9 Exercises in cointegration analysis
337(2)
16.9.1 Exercise 16.1
337(1)
16.9.2 Exercise 16.2
337(1)
16.9.3 Exercise 16.3
337(1)
16.9.4 Exercise 16.4
337(1)
16.9.5 Exercise 16.5
337(2)
17 Lessons in VARX Modelling and Trend/Cycle Dec.
339(35)
17.1 Lesson 17.1: Testing the long-run validity of PPP and IRP hypotheses using UK data
339(8)
17.2 Lesson 17.2: A macroeconomic model for Indonesia
347(4)
17.3 Lesson 17.3: Testing for over-identifying restrictions in the Indonesian model
351(5)
17.4 Lesson 17.4: Forecasting UK inflation
356(4)
17.5 Lesson 17.5: Permanent and transitory components of output and consumption in a small model of the US economy
360(6)
17.6 Lesson 17.6: The trend-cycle decomposition of interest rates
366(4)
17.7 Lesson 17.7: The US equity market and the UK economy
370(3)
17.8 Exercises in VARX modelling
373(1)
17.8.1 Exercise 17.1
373(1)
17.8.2 Exercise 17.2
373(1)
17.8.3 Exercise 17.3
373(1)
17.8.4 Exercise 17.4
373(1)
18 Lessons in SURE Estimation
374(13)
18.1 Lesson 18.1: A restricted bivariate VAR model of patents and output growth in the US
374(2)
18.2 Lesson 18.2: Estimation of Grunfeld-Griliches investment equations
376(2)
18.3 Lesson 18.3: Testing cross-equation restrictions after SURE estimation
378(1)
18.4 Lesson 18.4: Estimation of a static almost ideal demand system
379(3)
18.5 Lesson 18.5: Estimation of a New Keynesian three equation model
382(2)
18.6 Lesson 18.6: 2SLS and 3SLS estimation of an exactly identified system
384(2)
18.7 Exercises in SURE Estimation
386(1)
18.7.1 Exercise 18.1
386(1)
18.7.2 Exercise 18.2
386(1)
18.7.3 Exercise 18.3
386(1)
18.7.4 Exercise 18.4
386(1)
18.7.5 Exercise 18.5
386(1)
18.7.6 Exercise 18.6
386(1)
19 Lessons in Univariate GARCH Modelling
387(16)
19.1 Lesson 19.1: Testing for ARCH effects in montly $/£ exchange rates
387(2)
19.2 Lesson 19.2: Estimating GARCH models for monthly $/£ exchange rate
389(4)
19.3 Lesson 19.3: Estimating EGARCH models for monthly $/£ exchange rate
393(2)
19.4 Lesson 19.4: Forecasting volatility
395(1)
19.5 Lesson 19.5: Modelling volatility in daily exchange rates
396(2)
19.6 Lesson 19.6: Estimation of GARCH-in-mean models of US excess returns
398(3)
19.7 Exercises in GARCH modelling
401(2)
19.7.1 Exercise 19.1
401(1)
19.7.2 Exercise 19.2
402(1)
20 Lessons in Multivariate GARCH Modelling
403(14)
20.1 Lesson 20.1: Estimating DCC models for a portfolio of currency futures
403(4)
20.2 Lesson 20.2: Plotting the estimated conditional volatilities and correlations
407(2)
20.3 Lesson 20.3: Testing for linear restrictions
409(1)
20.4 Lesson 20.4: Testing the validity of the t-DCC model
410(3)
20.5 Lesson 20.5: Forecasting conditional correlations
413(1)
20.6 Lesson 20.6: MGARCH applied to a set of OLS residuals
414(2)
20.7 Exercises in Multivariate GARCH Estimation
416(1)
20.7.1 Exercise 20.1
416(1)
20.7.2 Exercise 20.2
416(1)
20.7.3 Exercise 20.3
416(1)
V Econometric Methods
417(124)
21 Econometrics of Single Equation Models
419(60)
21.1 Summary statistics and autocorrelation coefficients
419(2)
21.1.1 Box-Pierce and Ljung-Box tests
420(1)
21.2 Non-parametric estimation of the density function
421(1)
21.3 Estimation of spectral density
422(1)
21.4 Hodrick-Prescott (HP) filter
423(1)
21.5 Pesaran-Timmermann non-parametric test of predictive performance
424(1)
21.6 Ordinary least squares estimates
424(6)
21.6.1 Regression results
425(2)
21.6.2 Diagnostic test statistics (the OLS case)
427(3)
21.7 Statistical model selection criteria
430(2)
21.7.1 Akaike information criterion (AIC)
430(1)
21.7.2 Schwarz Bayesian criterion (SBC)
431(1)
21.7.3 Hannan and Quinn criterion (HQC)
431(1)
21.7.4 Consistency properties of the different model-selectio criteria
431(1)
21.8 Non-nested tests for linear regression models
432(3)
21.9 Non-nested tests for models with different transforamtions of the dependent variable
435(4)
21.9.1 The PE test statistic
435(1)
21.9.2 The Bera-McAleer test statistic
436(1)
21.9.3 The double-length regression test statistic
436(1)
21.9.4 The Cox non-nested statistics computed by simulation
437(2)
21.9.5 Sargan and Vuong's likelihood criteria
439(1)
21.10 The generalized instrumental variable method
439(3)
21.10.1 Two-stage least squares
440(1)
21.10.2 Generalized R2 for IV regressions
441(1)
21.10.3 Sargan's general mis-specification test
441(1)
21.10.4 Sargan's test of residual serial correlation for IV regressions
442(1)
21.11 Exact ML/AR estimators
442(5)
21.11.1 The AR(1) case
444(1)
21.11.2 The AR(2) case
444(1)
21.11.3 Covariance matrix of the exact ML estimators for the AR(1) and AR(2) options
445(1)
21.11.4 Adjusted residuals, R2, R2, and other statistics
445(2)
21.11.5 Log-likelihood ratio statistics for tests of residual serial correlation
447(1)
21.12 The Cochrane-Orcutt iterative method
447(2)
21.12.1 Covariance matrix of the CO estimators
449(1)
21.13 ML/AR estimators by the Gauss-Newton method
449(1)
21.13.1 AR(m) error process with zero restrictions
450(1)
21.14 The IV/AR estimation method
450(3)
21.14.1 Sargan's general mis-specification test in the case of the IV/AR option
451(1)
21.14.2 R2, R2, GR2, GR2, and other statistics: AR options
452(1)
21.15 Exact ML/MA estimators
453(3)
21.15.1 Covariance matrix of the unknown parameters in the MA option
455(1)
21.16 The IV/MA estimators
456(1)
21.16.1 R2, R2, GR2, GR2, and other statistics: MA options
456(1)
21.17 Recursive regressions
457(4)
21.17.1 The CUSUM test
458(1)
21.17.2 The CUSUM of squares test
458(1)
21.17.3 Recursive coefficients: the OLS option
458(1)
21.17.4 Standardized recursive residuals: the OLS option
459(1)
21.17.5 Recursive standard errors: the OLS option
459(1)
21.17.6 Recursive estimation: the IV option
459(1)
21.17.7 Adaptive coefficients in expectations formation models under incomplete learning
460(1)
21.17.8 Recursive predictions
460(1)
21.18 Phillips-Hansen fully modified OLS estimators
461(2)
21.18.1 Choice of lag windows w(s, m)
462(1)
21.18.2 Estimation of the variance matrix of the FM-OLS estimator
462(1)
21.19 Autoregressive distributed lag models
463(2)
21.20 Probit and Logit models
465(4)
21.20.1 Estimating and testing vector functions of β
467(1)
21.20.2 Fitted probability and fitted discrete values
467(1)
21.20.3 Measures of goodness of fit and related test statistics
468(1)
21.20.4 Forecasting with Probit/Logit models
468(1)
21.21 Non-linear estimation
469(2)
21.21.1 The non-linear least squares (NLS) method
470(1)
21.21.2 The non-linear instrumental variables (NL/IV) method
470(1)
21.22 Heteroscedasticity-consistent variance estimators
471(1)
21.23 Newey-West variance estimators
472(1)
21.24 Variance of vector function of estimators
473(1)
21.25 Wald statistic for testing linear and non-linear restrictions
474(1)
21.26 Univariate forecasts in regression models
474(5)
21.26.1 Univariate static forecasts
475(1)
21.26.2 Univariate dynamic forecasts
476(1)
21.26.3 Standard errors of univariate forecast errors: the OLS and IV options
476(1)
21.26.4 Forecasts based on non-linear models
477(1)
21.26.5 Measures of forecast accuracy
477(2)
22 Econometrics of Multiple Equation Models
479(46)
22.1 Seemingly unrelated regression equations (SURE)
480(2)
22.1.1 Maximum likelihood estimation
480(2)
22.2 Three-stage least squares
482(3)
22.2.1 Testing linear/non-linear restrictions
484(1)
22.2.2 LR statistic for testing whether Σ is diagonal
484(1)
22.3 System estimation subject to linear restrictions
485(2)
22.4 Augmented vector autoregressive models
487(3)
22.4.1 VAR order selection
488(1)
22.4.2 Testing the deletion of deterministic/exogenous variables
489(1)
22.4.3 Testing for block Granger non-causality
489(1)
22.5 Impulse response analysis
490(4)
22.5.1 Orthogonalized impulse responses
491(1)
22.5.2 Generalized impulse responses
492(2)
22.6 Forecast error variance decompositions
494(2)
22.6.1 Orthogonalized forecast error variance decomposition
494(1)
22.6.2 Generalized forecast error variance decomposition
494(2)
22.7 Cointegrating VAR
496(3)
22.7.1 Cointegrating relatious
497(2)
22.8 ML estimation and tests of cointegration
499(5)
22.8.1 Maximum eigenvalue statistic
502(1)
22.8.2 Trace statistic
502(1)
22.8.3 Model selection criteria for choosing the number of cointegrating relations
503(1)
22.9 Long-run structural modelling
504(7)
22.9.1 Identification of the cointegrating relations
504(1)
22.9.2 Estimation of the cointegrating relations under general linear restrictions
505(3)
22.9.3 Log-likelihood ratio statistics for tests of over-identifying restrictions on the cointegrating relations
508(1)
22.9.4 Impulse response analysis in cointegrating VAR models
509(1)
22.9.5 Impulse response functions of cointegrating relations
510(1)
22.9.6 Persistence profiles for cointegrating relations and speed of convergence to equilibrium
511(1)
22.10 VARX Models
511(7)
22.10.1 The structural VARX model
512(1)
22.10.2 The reduced form VARX model
513(1)
22.10.3 The cointegrated VAR model with I(1) exogenous variables
513(4)
22.10.4 Forecasting and impulse response analysis in VARX models
517(1)
22.11 Trend/cycle decomposition in VARs
518(3)
22.12 Principal components
521(2)
22.12.1 Selecting the number of PCs or factors
522(1)
22.13 Canonical correlations
523(2)
23 Econometrics of Volatility Models
525(16)
23.1 Univariate conditionally heteroscedastic models
525(7)
23.1.1 GARCH-in-mean models
525(2)
23.1.2 ML estimation with Gaussian errors
527(1)
23.1.3 ML estimation with Student's t-distributed errors
527(1)
23.1.4 Exponential GARCH-in-Mean models
528(1)
23.1.5 Absolute GARCH-in-Mean models
528(1)
23.1.6 Computational considerations
529(1)
23.1.7 Testing for ARCH (or GARCH) effects
529(1)
23.1.8 Residuals, DW, R2 and other statistics
529(1)
23.1.9 Forecasting with conditionally heteroscedastic models
530(2)
23.2 Multivariate conditionally heteroscedastic models
532(9)
23.2.1 Initialization, estimation and evaluation samples
534(1)
23.2.2 Maximum likelihood estimation
534(3)
23.2.3 Simple diagnostic tests of the DCC model
537(2)
23.2.4 Forecasting volatilities and conditional correlations
539(2)
Appendix A Size Limitations
541(2)
Appendix B Statistical Tables
543(4)
B.1 Upper and lower bound F-test and W-test critical values of Pesaran, Shin and Smith single-equation cointegration test
543(4)
References 547
Bahram Pesaran is currently a Research Consultant at Wadhwani Asset Management. He has also worked as a Research Analyst at Tudor Investment Corporation, The Bank of England, The National Institute of Economics and Social Research and The Confederation of British Industry.

Hashem Pesaran is Professor of Economics at the University of Cambridge, John Elliott Chair at the University of Southern California, and a Professorial Fellow of Trinity College, Cambridge. Previously he was the head of the Economic Research Department of the Central Bank of Iran, and Professor of Economics at the University of California at Los Angeles. Dr Pesaran is the founding editor of the Journal of Applied Econometrics and has served as a Vice President at the Tudor Investment Corporation. He is a Fellow of the Econometric Society and a Fellow of British Academy