Chapter 3 The Simple Linear Regression Model: Specification and Estimation

3.1 An Economic Model The simple regression function E ( y | x) = y| x = 1 + 2 x Slope of regression line 2 = E ( y | x) dE ( y | x) = x dx (3.1.2) (3.1.1)

denotes change in

Slide 3.1 Undergraduate Econometrics, 2nd Edition Chapter 3

3.2 An Econometric Model Assumptions of the Simple Linear Regression Model-I The average value of y, for each value of x, is given by the linear regression E ( y ) = 1 + 2 x For each value of x, the values of y are distributed about their mean value,

following probability distributions that all have the same variance, var( y ) = 2

Slide 3.2 Undergraduate Econometrics, 2nd Edition Chapter 3

The values of y are all uncorrelated, and have zero covariance, implying that

there is no linear association among them.

This assumption can be made stronger by assuming that the values of y are all statistically independent. The variable x is not random and must take at least two different values (optional) The values of y are normally distributed about their mean for each

value of x, y ~ N [(1 + 2 x), 2 ]

3.2.1

Introducing the Error Term

The random error term is e = y E ( y ) = y 1 2 x Rearranging gives y = 1 + 2 x + e y is dependent variable; x is independent or explanatory variable (3.2.2) (3.2.1)

Slide 3.4 Undergraduate Econometrics, 2nd Edition Chapter 3

Assumptions of the Simple Linear Regression Model-II SR1 SR2. SR3. SR4. y = 1 + 2 x + e E (e) = 0 E ( y ) = 1 + 2 x
var(e) = 2 = var( y ) cov(ei , e j ) = cov( yi , y j ) = 0

SR5. The variable x is not random and must take at least two different values. SR6. (optional) The values of e are normally distributed about their mean
e ~ N (0, 2 )

Slide 3.5 Undergraduate Econometrics, 2nd Edition Chapter 3

3.3 Estimating the Parameters for the Expenditure Relationship 3.3.1 The Least Squares Principle

The fitted regression line is yt = b1 + b2 xt The least squares residual et = yt yt = yt b1 b2 xt Any other fitted line
* yt* = b1* + b2 xt

(3.3.1)

(3.3.2)

(3.3.3)

Least squares line has smaller sum of squared residuals e = (y

y t )2 et*2 = ( yt yt* )2

Slide 3.6 Undergraduate Econometrics, 2nd Edition Chapter 3

 Least squares estimates are obtained by minimizing the sum of squares function S (1 , 2 ) = ( yt 1 2 xt ) 2
t =1 T

(3.3.4)

Math: Obtain partial derivatives S = 2T 1 2 yt + 2 xt 2 1 (3.3.5) S = 2 xt22 2 xt yt + 2 xt 1 2 Set derivatives to zero 2( yt Tb1 xt b2 ) = 0 (3.3.6) 2( xt yt xt b1 xt2b2 ) = 0

Slide 3.7 Undergraduate Econometrics, 2nd Edition Chapter 3

Rearranging equation 3.3.6 leads to two equations usually known as the normal equations, Tb1 + xt b2 = yt (3.3.7a)
t t

xb +x b = x y
t 1 2 t 2

(3.3.7b)

Formulas for least squares estimates b2 = T xt yt xt yt T x ( xt )

(3.3.8a) (3.3.8b)

b1 = y b2 x

Since these formulas work for any values of the sample data, they are the least squares estimators.

Slide 3.8 Undergraduate Econometrics, 2nd Edition Chapter 3

3.3.2

Estimates for the Food Expenditure Function T xt yt xt yt T x ( xt )

b2 =

(40)(3834936.497) (27920)(5212.520) (40)(21020623.02) (27920) 2

(3.3.9a)

= 0.1283 b1 = y b2 x = 130.313 (0.1282886)(698.0) = 40.7676 (3.3.9b)

A convenient way to report the values for b1 and b2 is to write out the estimated or fitted regression line: yt = 40.7676 + 0.1283 xt (3.3.10)

Slide 3.9 Undergraduate Econometrics, 2nd Edition Chapter 3

3.3.3

Interpreting the Estimates

The value b2 = 0.1283 is an estimate of 2, the amount by which weekly expenditure on food increases when weekly income increases by $1. Thus, we estimate that if income goes up by $100, weekly expenditure on food will increase by approximately $12.83. Strictly speaking, the intercept estimate b1 = 40.7676 is an estimate of the weekly amount spent on food for a family with zero income

Slide 3.10 Undergraduate Econometrics, 2nd Edition Chapter 3

3.3.3a

Elasticities

The income elasticity of demand is a useful way to characterize the responsiveness of consumer expenditure to changes in income. From microeconomic principles the elasticity of any variable y with respect to another variable x is = percentage change in y y / y y x = = percentage change in x x / x x y E ( y ) x (3.3.11)

In the linear economic model given by equation 3.1.1 we have shown that 2 = (3.3.12)

The elasticity of average expenditure with respect to income is = E ( y ) / E ( y ) E ( y ) x x = = 2 x / x x E ( y ) E ( y) (3.3.13)

Slide 3.11 Undergraduate Econometrics, 2nd Edition Chapter 3

A frequently used alternative is to report the elasticity at the point of the means ( x , y ) = (698.00, 130.31) since that is a representative point on the regression line. = b2 x 698.00 = 0.1283 = 0.687 y 130.31 (3.3.14)

3.3.3b

Prediction

Suppose that we wanted to predict weekly food expenditure for a household with a weekly income of $750. This prediction is carried out by substituting x = 750 into our estimated equation to obtain yt = 40.7676 + 0.1283 xt = 40.7676 + 0.1283(750) = $130.98 (3.3.15)

We predict that a household with a weekly income of $750 will spend $130.98 per week on food.

Slide 3.12 Undergraduate Econometrics, 2nd Edition Chapter 3

3.3.3c

Examining Computer Output

Dependent Variable: FOODEXP Method: Least Squares Sample: 1 40 Included observations: 40 Variable C INCOME R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat

Coefficient 40.76756 0.128289 0.317118 0.299148 37.80536 54311.33 -201.0297 2.370373

Std. Error 22.13865 0.030539 Mean dependent var S.D. dependent var Akaike info criterion Schwarz criterion F-statistic Prob(F-statistic)

t-Statistic 1.841465 4.200777

Prob. 0.0734 0.0002 130.3130 45.15857 10.15149 10.23593 17.64653 0.000155

Figure 3.10 EViews Regression Output

Slide 3.13 Undergraduate Econometrics, 2nd Edition Chapter 3

Dependent Variable: FOODEXP Analysis of Variance Source Model Error C Total Root MSE Dep Mean C.V. DF 1 38 39 Sum of Squares 25221.22299 54311.33145 79532.55444 Mean Square 25221.22299 1429.24556 R-square Adj R-sq F Value 17.647 Prob>F 0.0002

37.80536 130.31300 29.01120

0.3171 0.2991

Parameter Estimates Parameter Estimate 40.767556 0.128289 Standard Error 22.13865442 0.03053925 T for H0: Parameter=0 1.841 4.201

Variable INTERCEP INCOME

DF 1 1

Prob > |T| 0.0734 0.0002

Figure 3.11 SAS Regression Output

Slide 3.14 Undergraduate Econometrics, 2nd Edition Chapter 3

3.3.4

Other Economic Models

The log-log model ln( y ) = 1 + 2 ln( x) The derivative of ln(y) with respect to x is d [ln( y )] 1 dy = dx y dx

The derivative of 1 + 2 ln( x) with respect to x is d [1 + 2 ln( x)] 1 = 2 dx x Setting these two pieces equal to one another, and solving for 2 gives 2 = dy x = dx y (3.3.16)

Slide 3.15 Undergraduate Econometrics, 2nd Edition Chapter 3