------------------------------------------------------------------------
                        Running ESL and ESLWIN
------------------------------------------------------------------------
Soon you will be running things on the computer, estimating relations,
testing hypotheses, and so on.  Now is a good time to learn how to use
the program ESL/ESLWIN designed especially for this course.  Don't 
postpone this till the last minute because you won't be able to do the 
next assignment on time.

FROM YOUR HOME COMPUTER

If you bought the fourth edition of the book with the disk, follow the
directions on Page C-2 and install ESL (in MSDOS) and ESLWIN (in
Windows).  Then carry out the Practice Computer Sessions given below.
If you have an older edition of the book and hence no disk, then rent
or borrow a copy of the disk from a classmate or use one of the PC's
in Econ 100.

FROM PC LAB IN ECON 100

First close previous applications that may still be active.  If you
don't see the ESLWIN icon at the bottom right hand corner of
PROGRAM MANAGER, press ALT TAB until you see Program Manager.

PRACTICE COMPUTER SESSION IN BATCH MODE

All the examples in the book can be reproduced on the computer using
the Practice Sessions desribed in Appendix D (fourth edition) and
referred to in various places of the book.

In this session you will reproduce Practice Session 3.1 in the book
with ready-made ESL command files (known as "batch files") already
installed.  Double click (twice) the ESLWIN icon, then choose Estimate,
Batch mode, text-book practice session, and OK.  Follow screen directions
and select the input file ps3-1.esl and the data file data3-1.  View
output on screen and select File, Print, to print the output.  Bring
the output to class on Thursday, Feb.  12.

PRACTICE COMPUTER SESSION ON INTERACTIVE USAGE

Interactive usage is when you issue one command, look at the results, and
then decide what the next command should be.  As before, double click
the ESLWIN icon, then choose Estimate, Interactive, Text-book data set,
and OK.  Under filename select data3-2 and OK.  The data file will be read
and you will get the prompt ?.

First type:
       graph exphlth income ;  (the semicolon is important)
to graph expenditure on health against income.  Next type:
       ols exphlth 0 income ;

to estimate the model by OLS.  The first name is the dependent
variable.  The independent variables come next; 0 refers to the
constant term (which stands for the intercept).  The residuals of the
regression will be stored internally under the name uhat.  The command
      genr ut = uhat
saves them under the name ut for future reference.  Next generate the
"fitted" values of exphlth (observed values minus the residuals) with
the command
      genr yhat = exphlth - ut

[CAUTION: Do not use variable numbers in a genr command.  They will be
treated as numerical values and not as variable names which must be
typed in full.]

Type: 
      smpl  1 10           (this sets the sample range to 1-10)
then
      print  -o  exphlth  yhat  ut  ; 

to have the three series printed.  The flag -o (which is minus "oh"
with no space in between) tells ESL to print the values as in a table.

      quit

to exit ESL.  View the output BUT DO NO PRINT IT.
------------------------------------------------------------------------