*
* Example from page 172, section 6.2
* Effects of rescaling variables
*
cal 1988
open data table6-2.prn
data(format=prn,org=columns) 1988:1 1997:1
*
* Note that, in the regression output, RATS attempts to come up with a common
* format to show the regression coefficients and standard errors. This makes the
* results easier to read than if the decimal points didn't align. In the most
* extreme case (the millions on millions regression), the two coefficients
* differ by six orders of magnitude. In order to show the very large intercept in
* a common format with the small slope coefficient, only three decimal digits can
* be included. The coefficients are COMPUTED to the same precision (add a DISPLAY
* %BETA(2) instruction after the regression to see the slope coefficient to
* higher precision).
*
* Both in billions
*
linreg gpdibl
# constant gdpb
*
* Both in millions
*
linreg gpdim
# constant gdpm
*
* Dependent variable in billions, explanatory in millions
*
linreg gpdibl
# constant gdpm
*
* Dependent variable in millions, explanatory in billions
*
linreg gpdim
# constant gdpb