//  File MIT-Lab-17-01-03-01.txt. Edition 5/19/2012.


//  Lab Specs
//    Window Title
         Distribution_of_Error_Terms
//    List of Error Variances
         50. 500. 150. 500.
//    List of Heter Values
         -2.0 2.0 1.0 .0
//    List of Rho's
         -.9 .9 .3 .0
//    List of XECorr's
         -.9 .9 .3 .0


//  Problem Specs
//    Hetero visible (0 not visible, -1 visible)
//    Rho visible (0 not visible, -1 visible)
//    XECorr visible (0 not visible, -1 visible)
//    Graph visible (0 none, -1 et versus et-1, 1 e versus x)
//    Pause checked (0 not checked, -1 checked)
//    ErrVar value
//    Hetero Factor value
//    Rho value
//    XECorr value


` 0 -1 0 -1 -1 500 0 0 0
Objective: Illustrate the second ordinary least sqaures (OLS) standard premise, 
 the error term/error term independence premise: 
_
___The error terms are independent; their covariances are 0.
_
That is, knowing the value of one error term would not help us predict
 the value of any other error term.


` 0 -1 0 -1 -1 500 0 0 0
1. Click Start to run the first repetition of the experiment. Three histogram's appear
 at the top of the window; one histogram for each student's error term.
_
Also, a scatter diagram appears to the left. Notice that there are two points.
 The blue point represents the error terms of student 1 and 2 for the first repetition
 of the experiment. The red point represents the error terms of student 2 and 3
 for the first repetition.

`
Click Continue. Two different points now appear in the scatter diagram. The blue point
 representing the error terms of students 1 and 2 for the second repetition and The
 red point representing the error terms of students 2 and 3 for the second repetition.
_
Click Continue a few more times until that you are confident that you understand
 the scatter diagram.


` 0 -1 0 -1 0 500 0 0 0
A Rho value of 0 is specified in the Rho list; this means that no autocorrelation
 is present. Also, the Pause checkbox has been cleared. Click Continue and then 
 after many, many repetitions click Stop.
_
2. Focus on the scatter diagram. Does knowing the value of one error term help us
 predict the value of the next? That is, is one student's error term correlated
 with the next student's?
 

` 0 -1 0 -1 0 500 0 .9 0
Note that the Rho value has been increased from 0 to .9: autocorrelation is now present.
 Click Start and then after many, many repetitions click Stop.
_
3. Does knowing the value of one error term help us predict the value of the next?
 That is, is one student's error term correlated with the next student's?