Example 2.  Find the other "Least Squares Lines"  [Graphics:Images/LeastSqLineMod_gr_59.gif]  for the data points
[Graphics:Images/LeastSqLineMod_gr_60.gif].  
Use the subroutine Regression to find the line.
2 (a).  Use the computer to find the least squares lines  [Graphics:Images/LeastSqLineMod_gr_61.gif].   
2 (b).  Is it the same as the line we found in Example 1 ?   Why?

Solution 2.

First, enter the data points.

 

[Graphics:../Images/LeastSqLineMod_gr_62.gif]

Reverse the order of the coordinates of the points from  [Graphics:../Images/LeastSqLineMod_gr_63.gif]  to  [Graphics:../Images/LeastSqLineMod_gr_64.gif].

[Graphics:../Images/LeastSqLineMod_gr_65.gif]


[Graphics:../Images/LeastSqLineMod_gr_66.gif]

[Graphics:../Images/LeastSqLineMod_gr_67.gif]

Use Mathematica to find the "Least Squares Line"  [Graphics:../Images/LeastSqLineMod_gr_68.gif].  

[Graphics:../Images/LeastSqLineMod_gr_69.gif]



[Graphics:../Images/LeastSqLineMod_gr_70.gif]

[Graphics:../Images/LeastSqLineMod_gr_71.gif]
[Graphics:../Images/LeastSqLineMod_gr_72.gif]
[Graphics:../Images/LeastSqLineMod_gr_73.gif]
[Graphics:../Images/LeastSqLineMod_gr_74.gif]
[Graphics:../Images/LeastSqLineMod_gr_75.gif]


Plot the "Least Squares Line"  [Graphics:../Images/LeastSqLineMod_gr_76.gif]  using the  ParametricPlot  procedure.

[Graphics:../Images/LeastSqLineMod_gr_77.gif]


[Graphics:../Images/LeastSqLineMod_gr_78.gif]

[Graphics:../Images/LeastSqLineMod_gr_79.gif]

[Graphics:../Images/LeastSqLineMod_gr_80.gif]

[Graphics:../Images/LeastSqLineMod_gr_81.gif]

Compare the two "Least Squares Lines"  [Graphics:../Images/LeastSqLineMod_gr_82.gif].  

[Graphics:../Images/LeastSqLineMod_gr_83.gif]


[Graphics:../Images/LeastSqLineMod_gr_84.gif]

[Graphics:../Images/LeastSqLineMod_gr_85.gif]
[Graphics:../Images/LeastSqLineMod_gr_86.gif]


Are the two lines the same ? Solve the second formula for x in terms of y.

[Graphics:../Images/LeastSqLineMod_gr_87.gif]



[Graphics:../Images/LeastSqLineMod_gr_88.gif]
[Graphics:../Images/LeastSqLineMod_gr_89.gif]
[Graphics:../Images/LeastSqLineMod_gr_90.gif]

[Graphics:../Images/LeastSqLineMod_gr_91.gif]

[Graphics:../Images/LeastSqLineMod_gr_92.gif]

[Graphics:../Images/LeastSqLineMod_gr_93.gif]
[Graphics:../Images/LeastSqLineMod_gr_94.gif]


The two formulas are not the same.

The sum of the residual's squared for this example is:

 

[Graphics:../Images/LeastSqLineMod_gr_95.gif]

[Graphics:../Images/LeastSqLineMod_gr_96.gif]

Which can be compared with example 1.

 

[Graphics:../Images/LeastSqLineMod_gr_97.gif]

[Graphics:../Images/LeastSqLineMod_gr_98.gif]

Which "least squares fit" do you think is best ?

Question.  Are we comparing the "same thing" or is it apple's verses oranges?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(c) John H. Mathews 2004