13.9
a) BOXPLOTS
Boxplot of c1, c2, c3, c4, c5
155
150
Data
145
140
135
130
c1
c2
c3
c4
c5
Coatings 1, 2 and 5 may have the same mean, but their means seem to be different than coatings
3 and 4. Coatings 3 and 4 may have the same mean, which is smaller th

ISE 225
ENGINEERING STATISTICS
CLASS 13: MULTIPLE REGRESSION ANALYSIS
Detlof von Winterfeldt
Professor
University of Southern California
April 10, 2014
Housekeeping
Homework
HW 6 (ANOVA) was due today
Three more classes
April 17 (next week): Time seri

ISE 225 CLASS 4
ENGINEERING STATISTICS
Detlof von Winterfeldt
Professor
University of Southern California
Housekeeping
Homework 1 has been graded, overall a good result
Homework is due next week at 11:00 AM
I have asked for help in installing DATA Anal

ISE 225
ENGINEERING STATISTICS
CLASS 12: CHI-SQUARED TEST AND SIMPLE REGRESSION
Detlof von Winterfeldt
Professor
University of Southern California
April 3, 2014
Housekeeping
Finished grading Midterm
Maximum: 300 points
Mean:
Median:
25th:
75th:
HW

ISE 225 CLASS 7
ENGINEERING STATISTICS
Detlof von Winterfeldt
Professor, Price School and Viterbi School of Engineering
Housekeeping
I graded HW 2
Tough problem
Mean score was 32.5 out of 50
HW 3 was due today should have been easy
Wen will grade thi

ISE 225 CLASS 1
ENGINEERING STATISTICS
Detlof von Winterfeldt
Professor, Price School and Viterbi School of Engineering
Background - DvW
Born and grew up in Germany
Ph.D. in Mathematical Psychology, University of Michigan
1975-1978 - International Inst

ISE 225 CLASS 3
ENGINEERING STATISTICS I
Detlof von Winterfeldt
Professor, Price School and Viterbi School of Engineering
Housekeeping Notes
Homework 1 was due today before class
Our TA Wentoa will grade using my grading key (maximum = 50
points)
If yo

ISE 225 CLASS 11
ENGINEERING STATISTICS
Detlof von Winterfeldt
Professor, Price School and Viterbi School of Engineering
Housekeeping
Homeworks
HW 3 and 4 graded by TA
No HW 5 everyone gets 50 points!
Midterm
Solution has been posted and will be revi

ISE 225
ENGINEERING STATISTICS
CLASS 6: ESTIMATION AND INFERENCE
Detlof von Winterfeldt
Professor
University of Southern California
Housekeeping
Homework Assignments
Adjusted some of the grades upwards for HW 1
Will grade HW 2 on Saturday
HW 3 is post

ISE 225 CLASS 8
ENGINEERING STATISTICS
Detlof von Winterfeldt
Professor, Price School and Viterbi School of Engineering
Housekeeping
Homework
HW 3 is being graded by the TA
HW 4 is due next week
Today
A little bit more on Hypothesis Testing
Review

Exponential Smoothing with Trend
As we move toward medium-range
forecasts, trend becomes more important.
Incorporating a trend component into
exponentially smoothed forecasts is called
double exponential smoothing.
The estimate for the average and the

Time Series
Engineering Statistics I
4/2/2014
1
ISE 225, Dr. Smith
Elements of forecasting
Time frame (short, medium, long range)
Short~weeks; medium~months; long:~years
Patterns in the forecast
Trend: gradual up or down change in variable
(CPI)
Cycl

Nolan Dohnalek
ISE 225
require("PASWR2")
# Loading required package: PASWR2
# Warning: package 'PASWR2' was built under R version 3.2.3
# Loading required package: lattice
# Loading required package: ggplot2
# Warning: package 'ggplot2' was built under R

ISE 225 CLASS 2
ENGINEERING STATISTICS I
Detlof von Winterfeldt
Professor, Price School and Viterbi School of Engineering
Housekeeping Notes
Homework 1 is due next Thursday before class
Our TA Wentoa will grade using my grading key (maximum = 50
points)

Homework 8
Nolan Dohnalek
1. The outliers for both m1 and m2 are data points #59 and #48.
2.
Where w is price and y is highway
3. For MPG.City B1=- 1.0219
r-squared= 0.3535042
Standard Error= 7.809 on 91 degrees of freedom
For MPG.Highway B1= -1.0158
r-sq

Homework 6
11.98
a) Sample covariance:
Standard deviation of x and y:
Sample correlation:
= xy / xy = 7.4765/(0.6914)(15.625) = 0.6921
Slope b1:
b1 = Sxy/Sxx = 7.4765/0.478 = 15.6412
b) Although no simple curve will pass exactly through the points (there

HOMEWORK #4
a) Mean
Variance
St Dev
Skewness
10.468
0.409
0.640
0.246
One-Sample T: morning
Variable
morning
N
50
Mean
10.4675
StDev
0.6396
SE Mean
0.0904
95% CI
(10.2858, 10.6493)
We can conclude that the mean is 10.468 and with 95% confidence we conclud

1/27/2014
ConfidenceIntervals
Whatifthestandarddeviation,2
,unknown?
Mustestimatewithsamplestandarddeviation,
n
s:
( x X )2
ConfidenceIntervalsII
s
i 1
i
n 1
Thenusethetdistributionwithstatistic
EngineeringStatisticsI
t
X
s/ n
The1 percent2sidedCIonth

1/28/2014
HypothesisTesting
How should the hypotheses be set up?
Which goes in the null versus alternate?
There are two types of error, Type I () and
Type II ().
TestsofHypotheses
SetupGuide
The Type I error is the probability of rejecting Ho
when it

1/27/2014
ConfidenceIntervals
Whatifthestandarddeviation,2
,unknown?
Mustestimatewithsamplestandarddeviation,
n
s:
( x X )2
ConfidenceIntervalsII
s
i 1
i
n 1
Thenusethetdistributionwithstatistic
EngineeringStatisticsI
t
X
s/ n
The1 percent2sidedCIonth

2/3/2014
HypothesisTesting
1. Formulate null and alternate hypotheses
TestsofHypotheses,Part2
2. Set level of significance, , and sample size
(Alpha typically 0.05, 0.10, 0.01)
3. Select appropriate test (statistic) and rejection
rule
4. Collect data and