Applied economic forecasting using time series methods

Description

Economic forecasting is a key ingredient of decision making in the public and private sectors. This book provides the necessary tools to solve real-world forecasting problems using time-series methods. It targets undergraduate and graduate students as well as researchers in public and private institutions interested in applied economic forecasting. Cover Half title Applied Economic Forecasting using Time Series Methods Copyright page Contents Preface Part I Forecasting with the Linear Regression Model 1 The Baseline Linear Regression Model 1.1 Introduction 1.2 The basic specification 1.3 Parameter estimation 1.4 Measures of model fit 1.5 Constructing point forecasts 1.6 Interval and density forecasts 1.7 Parameter testing 1.8 Variable selection 1.9 Automated variable selection procedures 1.9.1 Forward selection (FWD) 1.9.2 Least angle regressions (LARS) 1.9.3 LASSO and elastic net estimator (NET). 1.10 Multicollinearity1.11 Example using simulated data 1.11.1 Data simulation procedure 1.12 Empirical examples 1.12.1 Forecasting Euro area GDP growth 1.12.2 Forecasting US GDP growth 1.13 A hint of dynamics 1.13.1 Revisiting GDP forecasting 1.13.2 Forecasting default risk 1.14 Concluding remarks 2 Model Mis-Specification 2.1 Introduction 2.2 Heteroskedastic and correlated errors 2.2.1 The Generalized Least Squares (GLS) estimator and the feasible GLS estimator 2.3 HAC estimators 2.4 Some tests for homoskedasticity and no correlation 2.5 Parameter instability. 2.5.1 The effects of parameter changes2.5.2 Simple tests for parameter changes 2.5.3 Recursive methods 2.5.4 Dummy variables 2.5.5 Multiple breaks 2.6 Measurement error and real-time data 2.7 Instrumental variables 2.8 Examples using simulated data 2.9 Empirical examples 2.9.1 Forecasting Euro area GDP growth 2.9.2 Forecasting US GDP growth 2.9.3 Default risk 2.10 Concluding remarks 3 The Dynamic Linear Regression Model 3.1 Introduction 3.2 Types of dynamic linear regression models 3.3 Estimation and testing 3.4 Model specification 3.5 Forecasting with dynamic models. 3.6 Examples with simulated data3.7 Empirical examples 3.7.1 Forecasting Euro area GDP growth 3.7.2 Forecasting US GDP growth 3.7.3 Default risk 3.8 Concluding remarks 4 Forecast Evaluation and Combination 4.1 Introduction 4.2 Unbiasedness and efficiency 4.3 Evaluation of fixed event forecasts 4.4 Tests of predictive accuracy 4.5 Forecast comparison tests 4.6 The combination of forecasts 4.7 Forecast encompassing 4.8 Evaluation and combination of density forecasts 4.8.1 Evaluation 4.8.2 Comparison 4.8.3 Combination 4.9 Examples using simulated data 4.10 Empirical examples. 4.10.1 Forecasting Euro area GDP growth4.10.2 Forecasting US GDP growth 4.10.3 Default risk 4.11 Concluding remarks Part II Forecasting with Time Series Models 5 Univariate Time Series Models 5.1 Introduction 5.2 Representation 5.2.1 Autoregressive processes 5.2.2 Moving average processes 5.2.3 ARMA processes 5.2.4 Integrated processes 5.2.5 ARIMA processes 5.3 Model specification 5.3.1 AC/PAC based specification 5.3.2 Testing based specification 5.3.3 Testing for ARCH 5.3.4 Specification with information criteria 5.4 Estimation 5.5 Unit root tests 5.6 Diagnostic checking.