Figures

The pdf files available here are the original figures submitted to Oxford University Press. During the production of the book the compositor made new figures so that there are differences between these images and those in the book. Nevertheless, we hope that these files will be useful to instructors for their lectures.

All of the files are also contained in a zipped file.

• 1.1 Wage versus Education
• 1.2 Wage versus Experience
• 1.3 Simple OLS Fit

---

• 2.1 Three Observations of y
• 2.2 Vector Representation of Data
• 2.3 Orthogonal Projection
• 2.4 Vector Decomposition
• 2.5 Ordinary Least Squares Projection in 2 Dimensions
• 2.6 Ordinary Least Squares Projection in 3 Dimensions
• 2.7 Decomposing mu-hat to get beta-hat
• 2.8 Decomposing mu-hat with Collinear RHS Variables

---

• 3.1 U. S. National Unemployment Rate and Fitted Seasonal Pattern
• 3.2 Seasonally Adjusted U. S. National Unemployment Rate
• 3.3 Unemployment Rate: Actual and Seasonal Forecast, 1993:12-1994:11
• 3.4 U. S. National Unemployment Rate and Dynamic Fit
• 3.5 Unemployment Rate: Actual Data and Dynamic Forecast, 1993:12--1994:11
• 3.6 Unemployment, Fitted Values, and Seasonal Component
• 3.7 Unemployment: Actual Data and Hybrid Forecast, 1993:12--1994:11
• 3.8 The Association Between Two Variables
• 3.9 The Association Between Three Variables
• 3.10 The Association Between Three Variables
• 3.11 The Association Between Three Variables
• 3.12 The Association Between Two Variables
• 3.13 Nonorthogonal Projection
• 3.14 Projection by P₁₂
• 3.15 Orthogonal RHS Variables

---

• 4.1 Unrestricted and Restricted Polynomial Distributed Lag Coefficients
• 4.2 RLS as a Projection of OLS
• 4.3 RLS as a Projection and Translation of OLS

---

• 6.1 Marginal Distribution of Experience
• 6.2 Marginal Distribution of Log-Wage
• 6.3 Joint Distribution of Experience and Log-Wage
• 6.4 Conditional Wage Distributions
• 6.5 Conditional Mean Given Experience
• 6.6 Frequency Distribution of Fitted Coefficients
• 6.7 Average OLS Quadratic versus Conditional Mean

---

• 7.1 The Scatter Plot and Variance Ellipse of Age and Experience
• 7.2 The Scatter Plot and Variance Ellipse of Education and Log-Wage
• 7.3 Variance Ellipse: Equal Versus Unequal Variances
• 7.4 Variance Ellipse: Noncovariance Versus Covariance
• 7.5 Variance Ellipse: Singular Covariance
• 7.6 The Variance Sphere of y for Two Observations
• 7.7 The Variance Sphere of y for Three Observations
• 7.8 The MMSE Predictor

---

• 8.1 Conditional Variance of Log Wage given Experience
• 8.2 Scatter Plot and Variance Ellipse of Experience Coefficients
• 8.3 Projection of Variance Sphere of y Onto Col(X)
• 8.4 Relationship Between V_mu and V_beta
• 8.5 Sphere and Rotated Sphere

---

• 9.1 Relative Efficiency
• 9.2 Scatter Plots for Large and Small Variances
• 9.3 Scatter Plots for Two Sample Sizes
• 9.4 Scatter Plots for Two Sample Variances of x₂
• 9.5 Increasing Collinearity
• 9.6 Forecast Variance in Simple Regression
• 9.7 Illustration of Relative Efficiency
• 9.8 Projection of V_y in 2 Dimensions
• 9.9 Projection of V_y in 3 Dimensions

---

• 10.1 95% Confidence Interval for Experience Coefficients
• 10.2 Bivariate Normal P.D.F.
• 10.3 Singular Bivariate Normal C.D.F.

---

• 11.1 95% Confidence Interval for Female and Nonwhite Coefficients
• 11.2 Distribution of Hypothesis Test Statistic
• 11.3 Joint versus Marginal Statistical Significance
• 11.4 Joint versus Marginal Statistical Significance
• 11.5 Distribution of Hypothesis Test Statistic Under Alternatives

---

• 13.1 The Laplace, Logistic, and Normal Distributions
• 13.2 Comparison of Tail Behavior
• 13.3 Relative Efficiency of Median versus Mean for t Distribution
• 13.4 F_3,N-K Distribution versus X²₃/3 Distribution
• 13.5 Approximate and Empirical C.D.F.'s
• 13.6 Approximate and Empirical C.D.F.'s
• 13.7 Approximate and Empirical C.D.F.'s
• 13.8 Approximate and Empirical C.D.F.'s
• 13.9 Sum of Absolute Residuals Function
• 13.10 F_{U_N}(z) versus F_U(z)

---

• 15.1 Nonuniform Convergence
• 15.2 Convergence of the MLE

---

• 16.1 Approximation of Sine by Quadratic and Cubic Polynomials
• 16.2 Line Search in a Two-Dimensional Parameter Space
• 16.3 Log-Likelihood Function in Step Length
• 16.4 Optimization by Steepest Ascent: Path on a Quadratic Function
• 16.5 Illustration of Convergence Criterion
• 16.6 Grid of Maximized Log-Likelihood Values in nu
• 16.7 Quadratic Approximation of L(sigma²)
• 16.8 Quadratic Approximation of L(log sigma²)
• 16.9 A Multi-Modal Log-Likelihood Function
• 16.10 View of Concentrated Log-Likelihood

---

• 17.1 Contours of the Concentrated Log-Likelihood Function
• 17.2 Contours of the Concentrated Log-Likelihood Function
• 17.3 The Relationship Among the Wald, LR, and Score Tests

---

• 18.1 Box Plots of OLS Fitted Residuals by Schooling Level
• 18.2 Box Plots of OLS Fitted Residuals by Experience Level
• 18.3 Heteroskedastic Variance Ellipsoid
• 18.4 Relative Efficiency of OLS and FWLS
• 18.5 Unbounded Log-Likelihood Function Allowing Linear Heteroskedasticity

---

• 19.1 Serial Correlation versus Omitted Explanatory Variables
• 19.2 Correlated Variance Ellipsoid

---

• 20.1 Errors in Variables

---

• 25.1 Cov[e_t, e_t+s] for Various ARMA Models
• 25.2 Estimates of the Autocorrelation Function

---

• 26.1 Fixed Supply and Demand Functions
• 26.2 Fixed Supply and Shifting Demand Functions

---

• 27.1 Binomial Dependent Variable
• 27.2 Alternative C.D.F.'s
• 27.3 Average Derivative versus Derivative at Average
• 27.4 Ordered Probability Model
• 27.5 Count Distributions

---

• 28.1 Labor Supply
• 28.2 Censored Regression
• 28.3 Censored C.D.F.
• 28.4 Censored P.D.F.
• 28.5 Censored P.D.F. with High Censoring Probability
• 28.6 Censored Mean for the Normal Distribution
• 28.7 Censored Variance for the Normal Distribution
• 28.8 Truncated P.D.F.
• 28.9 Truncated Mean for the Normal Distribution
• 28.10 Sample Selection P.D.F. for Various rho₀
• 28.11 Censored Mean
• 28.12 Censored Variance
• 28.13 Censored Weight
• 28.14 Truncated Mean Functions

---

• C.1 A Vector in Two Dimensions
• C.2 A Vector Sum in Two Dimensions
• C.3 A Scalar Product in Two Dimensions
• C.4 A Matrix as a Parallelogram
• C.5 Vector Addition of a Scalar Multiple of Another Column Vector
• C.6 Scalar Multiplication of a Column Vector
• C.7 Sum of Positive Determinants
• C.8 Sum of Positive and Negative Determinants

---

• D.1 The Normal and Student t Distributions
• D.2 Sequence of Densities for Average of Uniforms
• D.3 Sequence of Densities for Average of Exponentials