*   R. Carter Hill, William E. Griffiths and Guay C. Lim,  
*   Principles of Econometrics, Fourth Edition, Wiley, 2011.

* Working with the Simple Linear Regression Model
* Chapters 2, 3, and 4.

* Data set on food expenditure and weekly income from a 
* random sample of 40 households.
SAMPLE 1 40
READ (food.dat) / names

* The STAT command reports descriptive statistics (Table 2.1, p. 49)
STAT food income

* The OLS command reports a estimation results that include: 
*  - OLS parameter estimates (p. 53)
*  - the income elasticity evaluated at the sample means (p. 55)
*  - estimated standard errors (p. 36).
* Estimation results are summarized at top of page 56
* The OLS command also reports:
*  - the estimate of the error variance SIGMA**2 (p. 66)
*  - sum of squared residuals SSE (p. 66)
* The LIST option gives the predicted values and residuals (Table 2.3, p. 66).
*    The right portion of SHAZAM output from the LIST option shows
*    a crude residual plot.
* The PCOV option reports the estimated variances and covariances
*    of the least squares estimators (p. 66). 

* The OLS command also reports:
*  - t-test statistics for "tests of significance" and a p-value for
*    a 2-sided test.   
*  - the R-square (p. 138).

OLS food income / LIST PCOV 

* -------------------------------------------------------------
* The Effects of Scaling Data (Chapter 4.3.1, page 139)
* -------------------------------------------------------------
* Rescale income
GENR inc100=income*100
OLS food inc100

* -------------------------------------------------------------
* A Linear-Log Model (Chapter 4.3.3, page 144)
* -------------------------------------------------------------
* Get log-transformed income
GENR lninc=LOG(income)
* The LINLOG option on the OLS command calculates elasticities
* assuming log-transformed explanatory variables.
OLS food lninc / LINLOG

* -------------------------------------------------------------
* The Jarque-Bera test for normally distributed errors 
*      (Chapter 4.3.5, page 146)
* -------------------------------------------------------------
* On the OLS command the GF option reports tests for normality of the 
* errors including the Jarque-Bera test statistic and accompanying p-value.
OLS food income / GF
STOP