Index

Advanced time-series techniques

ARCH model, 50–51

ARMA models, 52–53

cointegration (see Cointegration)

GARCH model, 51–52

nonstationarity, 39–42

VAR model, 46–47

VEC model, 47–50

Augmented Dickey–Fuller test, 44–45

Autoregressive conditional heteroskedastic (ARCH) model, 50–51

Autoregressive Distributed Lag (ARDL) model, 39–41

Autoregressive moving average (ARMA) models, 52–53

Blundell-Bond System, 72

Bureau of Economic Analysis (BEA), 72, 105

Centers for Medicare & Medicaid Services (CMS), 106

Checking for balance, 67, 75–77

Choice models, 29

Chow test, 24

Cobb-Douglas production function, 6

Cointegration data analysis, 55–58

error term, 42

unit-root test augmented, 45

critical values, 46

error term, 45

OLS procedure, 46

Constant Returns to Scale (CRTS), 6

Cross sectional dummies, 22–23

Cumulative distribution function, 30

Data issue data processing, 106–107

data sources, 105–106

missing observations, 107–108

Data scaling effect, 5

Dickey–Fuller test, 43

augmented, 44–45

constant term, 43

LM test, 44

Tau-statistics, 43–44

Differences-in-differences (DD) approach

data analyses, 94–96

relaxing DD assumptions, 92

Tohoku earthquake, 89–91

Durable expenditures (DUR), 4

Econometric model, 102

Empirical research paper writing abstract, 100

conclusion, 104

econometric model, 102

estimation method and data, 102–103

introduction, 100–101

literature review, 101–102

references, 104–105

results, 103–104

Endogeneity

correction, 11–14

data analyses, 15–19

detection, 9–11

lagged-dependent model, 8

measurement error, 8–9

omitted variables, 7–8

random-regressor problem, 6–7

simultaneous equation bias, 9

Fixed effect, 72–73

Foreign direct investment (FDI), 26–27

Fuzzy regression discontinuity (RD), 87–89

Generalized ARCH (GARCH) model, 51–52

Generalized least squares (GLS) estimators, 30, 103

Generalized method of moments (GMM) procedure, 72

Grade point average (GPA), 65

Graduated licensing system, 87

Hausman test, 10

Instrumental variable (IV) estimators, 11–12

Integrated Public Use Microdata Series (IPUMS), 106

Interaction of continuous variables, 25–26

of qualitative factors, 23–24

International Monetary Fund (IMF), 106

Irrelevant variables, 4–5

Lagged-dependent model, 8

Lagrange multiplier (LM) tests, 44

Law of Large Number (LLN), 66–67

Least Squares Dummy Variable (LSDV) technique, 23

Linear probability model, 29–30

Linear spline models, 33

Logarithmic models, 108

Logit model, 30–31

Log-linear models, 27–29

Method of Moments (MM), 11–12

Minimum driving age (MDA), 87–88

Missing observations, 108–109

Modeling issues data scaling effect, 5

irrelevant variables, 4–5

nonsample information, 6

omitted variables, 3–4

Multinomial logit model, 30

Multiplicative dummy variables, 22–23

National Bureau of Economic Research (NBER), 105

Nonlinear models

choice models, 29

continuous variable interaction, 25–26

data analysis, 33–37

foreign direct investment, 26–27

linear probability model, 29–30

log-linear models, 27–29

piecewise and linear spline models, 32–33

polynomial model, 31–32

probit and logit models, 30–31

Nonsample information, 6, 14–15

Nonstationarity ARDL model, 39–41

trend stationarity, 41–42

Omitted variables, 3–4

Ordinary Least Squares (OLS) procedure, 4, 7, 36, 40, 42, 46–49, 103

Panel Survey of Income Dynamics (PSID), 106

Piecewise-downward approach, 4–5

Piecewise model, 32–33

Pitfalls in empirical research being careless on literature review, 109

failing to ask questions, 109

giving up too early, 110–111

perfect dataset, 109–110

too complicated model, 110

Polynomial model, 31–32

Probability model, 29, 35, 36

Probit model, 30–31

Quadratic spline regression, 33

Randomized-trial experiments, 66–68

Random-regressor problem, 6–7

Random walk process, 39–40

Regression discontinuity (RD) data analyses, 92–94

fuzzy RD, 87–88

school system, 83

specifics, 86–87

Tohoku earthquake and tsunami direct damage, 85, 86

emergency events database website, 85

Japanese tourists, 84

multiple-regression inference, 84

Root Mean Square Error (RMSE), 14

Running variable, 85

Scaling the data, 5

Selection bias apple-to-apple comparing process, 63–64

balanced data, 67, 68

community college vs. university Cobb–Douglas production function, 71

conservative approach, 70

earning by education attainment, 69

earning comparison, 70

new growth model, 71

nonregression approach, 69

selective-group dummies, 75

VIF tests, 72

data analyses, 75–78

Law of Large Numbers, 66–67

multiple regression method, 68

randomized results, 67–68

randomized-trial experiments, 66

two-student scenario, 65

Selective-group dummy, 73, 75, 77–79

Slope dummy variables cross sectional dummies, 22–23

interaction of qualitative factors, 23–24

time control, 25

two regressions equivalence testing, 24–25

Tau-statistics, 43–44

Treatment variable, 85, 90–91

Trend stationarity, 41–42

Two-stage least squares (2SLS) estimation

for fuzzy RD, 88–89

single equation model, 12–13

system of equations, 13–14

Unit-root tests cointegration test augmented, 45

critical values, 46

error term, 45

OLS procedure, 46

data analysis, 53–54

Dickey–Fuller test augmented, 44–45

constant term, 43

LM test, 44

Tau-statistics, 43–44

τ-critical values, 44

Variance Inflation Factors (VIF) tests, 72

Vector autoregressive (VAR) model, 46–47

Vector error correction (VEC) model advantages, 47

error term, 49

OLS procedure, 47–48

World Bank (WB), 106

World Trade Organization (WTO), 106