MDM4U 1.

5 Linear Regression ( technology)

The Correlation Coefficient !

1
The correlation coefficient, r, is a number between ______and 1
_______that is an indicator of both

linear
the strength and direction of a _________ relationship between two variables. A value of r = 0
indicates no correlation, while r = 1 or r = -1 indicates a perfect positive or negative correlation.

perfect perfectt

nå ( xy ) - (å x )(å y )
r=
[nå x 2
- (å x )
2
] [nå y 2
- (å y )
2
]
The Coefficient of Determination !!

1 0
The coefficient of determination ", is a number between ________ and __________ that is useful
because it gives the ______________
proportion of the variance (fluctuation) of one variable that is predictable
from the other variable. It is a measure that allows us to determine how certain one can be in making
predictions from a certain model/graph.

The coefficient of determination is a measure of how well the regression line (line of best fit)
represents the data. If the regression line passes exactly through every point on the scatter plot, it
would be able to explain all of the variation. Thefurther
__________ the line is away from the points, the
less it is able to explain.

85
For example, if r = 0.922, then r2 = 0.850, which means that ___________of the total variation in y
can be explained by the 1in retation
___________between x and y (as described by the regression equation). The

15
other _____________of the total variation in y remains unexplained.

å (y - y )
2
est
r2 =
å (y - y )
 MDM4U 1.5 Linear Regression ( technology)

Regression Line (Line of Best Fit)

A regression line is a line that describes how a dependent (response) variable # changes as
an independent (explanatory) variable $ changes. We often use a regression line to
________the value of # for a given value of $.
predict
We use the equation: #! ~&$! + (, where
a = rate of change or slope of the line
b = y intercept Age Annual
Income
Using DESMOS to determine the line of best fit: (Thousands
$)
1. Enter the data in the table of values in DESMOS by adding a new item 33 33
25 31
19 18
44 52
50 56
54 60
38 44
29 35

2. Click on Table of values and type in the values

3. In the next line type in this equation – activate the DESMOS keyboard by clicking in the bottom left
corner and choose ABC . You must use the ab to insert the equation : !1 ~#$1 + &

atb.sc notin
j
or
Desmos but
encounter
you might
4. Desmos will calculate the r and r2 values: which
this notation
is equivalent

5. What do the r values tell you about the correlation?

throng
 MDM4U 1.5 Linear Regression ( technology)

Residual Values

A residual is the difference between an observed

value of # and the value predicted by the regression
line (#). The residual value tells us how far off the
linear regression’s prediction is at a given point.
Residual = observed # – predicted #
=# −#

Example 2: Using the regression equation from example 1, find and interpret the residual for a truck
that had 70583 miles driven and a price of $21994.

IT
Solution:

Note: If the regression model is a good fit, the residuals should be fairly ______________ , and
there should be no noticeable pattern. _____________ residuals or a _____________are indicators
that another model may be more appropriate.

Example 3: Sketch the residual plot for the following graph and comment about what it tells you.

teased