Académique Documents
Professionnel Documents
Culture Documents
Walter R. Paczkowski
Rutgers University
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 1
Chapter Contents
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 2
4.1
Least Squares Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 3
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 4
4.1
Least Squares
Prediction
Eq. 4.1 y0 β1 β2 x0 e0
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 5
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 6
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 7
4.1
Least Squares
Prediction Figure 4.1 A point prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 8
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 9
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 10
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 11
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 12
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 13
4.1
Least Squares
Prediction
In practice we use
1
2
x - x
var f σˆ 2 1 0
N xi - x
2
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 14
4.1
Least Squares
Prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 15
4.1
Least Squares
Prediction
Figure 4.2 Point and interval prediction
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 16
4.1
Least Squares
Prediction
4.1.1
Prediction in the
Food Expenditure
For our food expenditure problem, we have:
Model
var f ˆ 1
2
N xi x
2
ˆ 2
ˆ 2
ˆ 2 x0 x
2
i
2
N x x
ˆ 2
ˆ x0 x var b2
2 2
N
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 17
4.1
Least Squares
Prediction
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 18
4.1
Least Squares
Prediction
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 19
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 20
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
To develop a measure of the variation in yi that is
explained by the model, we begin by separating yi
into its explainable and unexplainable
components.
Eq. 4.8 yi E yi ei
– E(yi) is the explainable or systematic part
– ei is the random, unsystematic and
unexplainable component
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 21
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 22
4.2
Measuring
Goodness-of-fit
Figure 4.3 Explained and unexplained components of yi
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 23
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
s 2
yˆ y i
N 1
y
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 24
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
yi y yˆi y eˆi2
2 2
Eq. 4.11
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 25
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 26
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Specifically:
y y total sum of squares SST
2
i
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 27
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 28
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 29
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
We can see that:
Model
– The closer R2 is to 1, the closer the sample
values yi are to the fitted regression equation
– If R2 = 1, then all the sample data fall exactly
on the fitted least squares line, so SSE = 0, and
the model fits the data ‘‘perfectly’’
– If the sample data for y and x are uncorrelated
and show no linear association, then the least
squares fitted line is ‘‘horizontal,’’ and identical
to y, so that SSR = 0 and R2 = 0
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 30
4.2
Measuring
Goodness-of-fit
4.1.1
Prediction in the
Food Expenditure
Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 31
4.2
Measuring
Goodness-of-fit
4.2.1
Correlation
Analysis
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 32
4.2
Measuring
Goodness-of-fit
4.2.1
Correlation Substituting sample values, as get the sample
Analysis
correlation coefficient:
sxy
rxy
sx s y
where:
sxy xi x yi y N 1
x x N 1
2
sx i
y y N 1
2
sy i
4.2.2
Correlation
Analysis and R2
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 34
4.2
Measuring
Goodness-of-fit
4.2.3
The Food
Expenditure
Example
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 35
4.2
Measuring
Goodness-of-fit
4.2.3
The Food
Expenditure
Example
Therefore:
SSE
R2 1
SST
304505.176
1
495132.160
0.385
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 36
4.2
Measuring
Goodness-of-fit
4.2.3
The Food
Expenditure
The sample correlation between the y and x
Example
sample values is:
sxy
rxy
sx s y
478.75
6.848112.675
0.62
– As expected:
rxy2 0.622 0.385 R2
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 37
4.2
Measuring
Goodness-of-fit
4.2.4
Reporting the
Results
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 38
4.2
Measuring
Goodness-of-fit
4.2.4
Reporting the
Results
For our food expenditure example, we might have:
FOOD_EXP = weekly food expenditure by a
household of size 3, in dollars
INCOME = weekly household income, in $100 units
And:
FOOD _ EXP 83.42 10.21INCOME R 2 0.385
se 43.41 2.09
***
where
* indicates significant at the 10% level
** indicates significant at the 5% level
*** indicates significant at the 1% level
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 39
4.3
Modeling Issues
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 40
4.3
Modeling Issues
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 41
4.3
Modeling Issues
4.3.1
The Effects of
Scaling the Data
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 42
4.3
Modeling Issues
4.3.1
The Effects of If we had estimated the regression using income in
Scaling the Data
dollars, the results would have been:
FOOD _ EXP 83.42 0.1021INCOME $ R 2 0.385
se 43.41 0.0209
***
4.3.1
The Effects of
Scaling the Data
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 44
4.3
Modeling Issues
4.3.1
The Effects of
Scaling the Data
4.3.1
The Effects of
Scaling the Data
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 46
4.3
Modeling Issues
4.3.2
Choosing a
Functional Form
4.3.2
Choosing a
Functional Form
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 48
4.3
Modeling Issues Figure 4.4 A nonlinear relationship between food expenditure and income
4.3.2
Choosing a
Functional Form
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 49
4.3
Modeling Issues
4.3.2
Choosing a By transforming the variables y and x we can represent
Functional Form
many curved, nonlinear relationships and still use the
linear regression model
– Choosing an algebraic form for the relationship
means choosing transformations of the original
variables
– The most common are:
• Power: If x is a variable, then xp means raising
the variable to the power p
– Quadratic (x2)
– Cubic (x3)
• Natural logarithm: If x is a variable, then its
natural logarithm is ln(x)
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 50
4.3
Modeling Issues Figure 4.5 Alternative functional forms
4.3.2
Choosing a
Functional Form
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 51
4.3
Modeling Issues
4.3.2
Choosing a
Functional Form
Summary of three configurations:
1. In the log-log model both the dependent and
independent variables are transformed by the
‘‘natural’’ logarithm
• The parameter β2 is the elasticity of y with
respect to x
2. In the log-linear model only the dependent
variable is transformed by the logarithm
3. In the linear-log model the variable x is
transformed by the natural logarithm
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 52
4.3
Modeling Issues
4.3.2
Choosing a
Functional Form
y β2
100 x x 100
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 53
4.3 Table 4.1 Some Useful Functions, their Derivatives, Elasticities and Other
Modeling Issues
Interpretation
4.3.2
Choosing a
Functional Form
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 54
4.3
Modeling Issues
4.3.3
A Log-linear Food
Expenditure Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 55
4.3
Modeling Issues
4.3.3
A Log-linear Food
Expenditure Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 56
4.3
Modeling Issues
4.3.3
A Log-linear Food
Expenditure Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 57
4.3
Modeling Issues
4.3.3
A Log-linear Food
Expenditure Model
For a household with $1,000 weekly income, we
estimate that the household will spend an additional
$13.22 on food from an additional $100 income
– Whereas we estimate that a household with $2,000
per week income will spend an additional $6.61
from an additional $100 income
– The marginal effect of income on food expenditure
is smaller at higher levels of income
• This is a change from the linear, straight-line
relationship we originally estimated, in which
the marginal effect of a change in income of
$100 was $10.21 for all levels of income
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 58
4.3
Modeling Issues
4.3.3
A Log-linear Food
Expenditure Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 59
4.3
Modeling Issues
Figure 4.6 The fitted linear-log model
4.3.3
A Log-linear Food
Expenditure Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 60
4.3
Modeling Issues GUIDELINES FOR CHOOSING A FUNCTIONAL FORM
4.3.3
A Log-linear Food
Expenditure Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 61
4.3
Modeling Issues
4.3.4
Using Diagnostic
Residual Plots
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 62
4.3
Modeling Issues Figure 4.7 Randomly scattered residuals
4.3.4
Using Diagnostic
Residual Plots
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 63
4.3
Modeling Issues Figure 4.8 Residuals from linear-log food expenditure model
4.3.4a
Homoskedastic
Residual Plot
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 64
4.3
Modeling Issues
4.3.4b
Detecting Model
Specification Errors
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 65
4.3
Modeling Issues Figure 4.9 Least squares residuals from a linear equation fit to quadratic data
4.3.4b
Detecting Model
Specification Errors
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 66
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
Distributed?
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 67
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
Distributed?
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 68
4.3 Figure 4.10 EViews output: residuals histogram and summary statistics for
Modeling Issues
food expenditure
4.3.5
Are the Regression
Errors Normally
Distributed?
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 69
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
The Jarque–Bera test for normality is based on
Distributed? two measures, skewness and kurtosis
– Skewness refers to how symmetric the residuals
are around zero
• Perfectly symmetric residuals will have a
skewness of zero
• The skewness value for the food expenditure
residuals is -0.097
– Kurtosis refers to the ‘‘peakedness’’ of the
distribution.
• For a normal distribution the kurtosis value
is 3
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 70
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
Distributed?
JB S
6 4
where
N = sample size
S = skewness
K = kurtosis
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 71
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
Distributed?
When the residuals are normally distributed, the
Jarque–Bera statistic has a chi-squared distribution
with two degrees of freedom
– We reject the hypothesis of normally distributed
errors if a calculated value of the statistic
exceeds a critical value selected from the chi-
squared distribution with two degrees of
freedom
• The 5% critical value from a χ2-distribution
with two degrees of freedom is 5.99, and the
1% critical value is 9.21
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 72
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
Distributed?
For the food expenditure example, the Jarque–
Bera statistic is:
40
JB 0.097
2 2.99 3
2
0.063
6
4
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 73
4.3
Modeling Issues
4.3.5
Are the Regression
Errors Normally
Distributed?
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 74
4.4
Polynomial Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 75
4.4
Polynomial Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 76
4.4
Polynomial Models
4.4.1
Quadratic and
Cubic Equations
y a 0 a1 x a 2 x 2
y a 0 a1 x a 2 x a 3 x 2 3
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 77
4.4
Polynomial Models Figure 4.11 Scatter plot of wheat yield over time
4.4.2
An Empirical
Example
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 78
4.4
Polynomial Models
4.4.2
An Empirical
Example
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 79
4.4
Polynomial Models Figure 4.12 Residuals from a linear yield equation
4.4.2
An Empirical
Example
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 80
4.4
Polynomial Models
4.4.2
An Empirical
Example
4.4.2
An Empirical
Example
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 82
4.5
Log-linear Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 83
4.5
Log-linear Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 85
4.5
Log-linear Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 86
4.5
Log-linear Models
4.5.1
A Growth Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 87
4.5
Log-linear Models
4.5.1
A Growth Model
ln YIELDt 0.3434 0.0178t
se 0.0584 0.0021
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 88
4.5
Log-linear Models
4.5.1
A Growth Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 89
4.5
Log-linear Models
4.5.2
The Wage
Equation
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 90
4.5
Log-linear Models
4.5.2
The Wage
Equation
ln WAGE 1.6094 0.0904 EDUC
se 0.0864 0.0061
– An additional year of education increases the
wage rate by approximately 9%
• A 95% interval estimate for the value of an
additional year of education is 7.8% to
10.2%
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 91
4.5
Log-linear Models
4.5.3
Prediction in the
Log-linear Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 92
4.5
Log-linear Models
4.5.3
Prediction in the
Log-linear Model
A natural choice for prediction is:
yˆ n exp ln y exp b1 b2 x
yˆc E y exp b1 b2 x σˆ 2 yˆ ne 2
σˆ 2 2
4.5.3
Prediction in the
Log-linear Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 94
4.5
Log-linear Models
4.5.3
Prediction in the
Log-linear Model
yˆ n exp ln y exp 2.6943 14.7958
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 95
4.5
Log-linear Models
4.5.3
Prediction in the
Log-linear Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 96
4.5
Log-linear Models FIGURE 4.14 The natural and corrected predictors of wage
4.5.3
Prediction in the
Log-linear Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 97
4.5
Log-linear Models
4.5.4
A Generalized R2
Measure
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 98
4.5
Log-linear Models
4.5.4
A Generalized R2
Measure
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 99
4.5
Log-linear Models
4.5.5
Prediction
Intervals in the
Log-linear Model
c
exp ln y t se f , exp ln y t se f
c
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 100
4.5
Log-linear Models
4.5.5
Prediction
Intervals in the
Log-linear Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 101
4.5
Log-linear Models FIGURE 4.15 The 95% prediction interval for wage
4.5.5
Prediction
Intervals in the
Log-linear Model
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 102
4.6
Log-log Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 103
4.6
Log-log Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 104
4.6
Log-log Models
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 105
4.6
Log-log Models FIGURE 4.16 Quantity and Price of Chicken
4.6.1
A Log-log Poultry
Demand Equation
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 106
4.6
Log-log Models
4.6.1
A Log-log Poultry
Demand Equation
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 107
4.6
Log-log Models
4.6.1
A Log-log Poultry
Demand Equation
Using the estimated error variance ˆ 2 = 0.0139,
the corrected predictor is:
ˆ ˆ
Qc Qn eˆ 2 2
exp ln Q e ˆ 2 2
2
g
R corr Q, Qˆ c 0.939
2 2
0.8817
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 108
Key Words
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 109
Keywords
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 110
Appendices
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 111
4A Development of a Prediction Interval
4B The Sum of Squares Decomposition
4C The Log-Normal Distribution
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 112
4A.1
Development of a
Prediction Interval
We know that:
N xi x
i
i
2 0 2 2
x x x x
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 113
4A.1
Development of a
Prediction Interval
2 x
2 Nx 2
x0 2
2 x0 2
2
xi x xi x N xi x
2 2
xi2 Nx 2 x02 2 x0 x x 2
2
2
i
N x x 2
i x x
x x 2
x0 x
2
2 i
2
i
N x x 2
i x x
2 1
x0 x
2
2
N i x x
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 114
4A.1
Development of a
Prediction Interval
Therefore:
f y0 yˆ0
~ t N 2
var f se f
Eq. 4A.1
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 115
4A.1
Development of a
Prediction Interval
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 116
4A.2
Sum of Squares
Decomposition
iy y 2
iˆ
y y 2
eˆi2 2 yˆ i y eˆi
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 117
4A.2
Sum of Squares
Decomposition
That is:
eˆ eˆi N 0
x ˆ
e x y b b x
i i i i 1 2 i i i 1 i 2 i 0
x y b x b x 2
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 119
4A.2
Sum of Squares
Decomposition
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 120
4A.3
The Log-Normal
Distribution
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 121
4A.3
The Log-Normal
Distribution
Ew e 2 2
and
var w e 2 2
e 2
1
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 122
4A.3
The Log-Normal
Distribution
E yi E e E e
β1 β 2 xi ei β1 β 2 xi ei
e
e β1 β 2 xi
E e ei
β1 β 2 xi σ 2 2
e e
β1 β 2 xi σ 2 2
e
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 123
4A.3
The Log-Normal
Distribution
E yi e b1 b2 xi ˆ 2 2
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 124
4A.3
The Log-Normal
Distribution
Therefore:
b2 var b2 2
rˆ e 1
where
var b2 ˆ xi x
2
Principles of Econometrics, 4th Edition Chapter 4: Prediction, Goodness-of-fit, and Modeling Issues Page 125