STAT 504 Lesson5: Simple Linear Regression

Simple Linear Regression Water example

At a particular production plant(industry is not specified so Ill assume theyre an

industrial bread bakery), they want to study how much water their facility is using.
It is probably true that the more bread they make in a given month, the more
water they end up using.

Data file from: Draper, N.R., and Smith, H. (1981). Applied Regression Analysis
(2nded.) New York: Wiley, page 352.

Available online at OzDASL, http://www.statsci.org/data/general/water.txt

Regression Analysis: Water versus Production

The regression equation is

Water = 2273 + 0.0799 Production

Predictor Coef SE Coef T P

Constant 2273.1 338.7 6.71 0.000
Production 0.07989 0.02538 3.15 0.007

S = 358.000 R-Sq = 39.8% R-Sq(adj) = 35.8%

Analysis of Variance

Source DF SS MS F P
Regression 1 1270172 1270172 9.91 0.007
Residual Error 15 1922459 128164
Total 16 3192632

Unusual Observations

Obs Production Water Fit SE Fit Residual St Resid

16 18573 4488.0 3756.9 168.1 731.1 2.31R

R denotes an observation with a large standardized residual.

 Fitted Line Plot
Water = 2273 + 0.07989 Production
S 358.000
4500
R-Sq 39.8%
R-Sq(adj) 35.8%

4000
r
e
t
a
W
3500

3000

5000 7500 10000 12500 15000 17500 20000

Production

Residual Plots for Water

Normal Probability Plot of the Residuals Residuals Versus the Fitted Values
99
600
90
t l
n a
e u300
c d
is
r
e
50 e
P R 0

10 -300

1 -600
-1000 -500 0 500 1000 3000 3300 3600 3900
Residual Fitted Value

Histogram of the Residuals Residuals Versus the Order of the Data

4.8
600
y
c l
n a
3.6
e u300
u d
i
q s
e e
r
2.4 R 0
F
1.2 -300

0.0 -600
-600 -400 -200 0 200 400 600 800 2 4 6 8 10 12 14 16
Residual Observation Order