Table of Contents
Advertisement
6. Analyze the data: Press
(Regression). You will see:
r
~
C22: [W151]
FEHL
1
-Data
0
Z
tercept
413.49488215
=
d Err of ¥Y Est
843.99857721
i
Squared
}
8.9731872975
=
. of Observations
11
=
ope
8.8040123457
Ee]
d Err of Coef.
0.4688880934
x
Predicted Value
413.48488215
alue
\_
_/
The macro has calculated a two-variable linear regression:
Intercept is the y-axis intercept for the regression line.
Std Err of Y Est isthe standard error of the estimated y values.
R squaredis the reliability of the regression (between 0 and 1).
Slope is the slope for the independent variable.
Std Err of Coef isthe standard error of the x coefficient (the slope).
Y¥ Predicted Value is the predicted value for the y-variable for the
given value of the x-variable. Since no x-value is given yet, x is assumed to be
equal to zero.
# Yalue is a value for the independent variable (i.e. floor space) that you
enter in order to predict a corresponding y-value (i.e. rental income), based on
the current regression.
7. Enter a different x-value. How much income can you expect from a building
with 2985 square feet of floor space? Press (2]9)]8]5)[ENTER). The predicted y-
value is 26693 .381 734 or about $26, 690 of rental income.
8. Plot the regression line. Press (ALT)}{P). The plot is good enough to give you a
visual estimate for how good the fit is. Of course, the r-squared value (see
above) gives you a calculated estimate of the same thing.
9. Ifyou want to save the data, press (MENU), (Flile, (Slave, type in a new name, and
press (ENTER).
Statistics and Databases
69
Advertisement
Table of Contents
