SPSS GUIDE
by Jafri
Solved Examples of Different Statistical
Techniques from Text Books
β
Σ
λ
α
Σ
σ
γ
χ
π
MUHAMMAD KAZIM JAFRI, 2009
m.kazimjafri9@gmail.com
Department of Statistics
University of Sindh Jamshoro
0|Page
Contents
Data Representation
Topic # 1- Pie Chart
Topic # 2: Dot Plot
Topic # 3: Box Plot
Topic # 4: Cumulative Frequency Polygon
Topic # 5: Scatter Diagram
Topic # 6: Scatter Plot Matrix
Descriptive Statistics
Topic # 7: Confidence Interval of Mean
Topic # 8: Harmonic Mean
Topic # 9: Trimmed Mean
Topic # 10: Standard Scores
Topic # 11: Standard Error of Mean
Probability Distributions
Topic # 12: Geometric Distribution
Topic # 13: Exponential Distribution
Topic # 14: Normal Distribution
Topic # 15: Uniform Distribution
Correlation
Topic # 16: Correlation for Grouped data
Topic # 17: Rank Correlation
Topic # 18: Rank Correlation for Tied Ranks
Topic # 19: Covariance
Regression Analysis
Topic # 20: Test of Linearity
Topic # 21: Mean Prediction
Topic # 22: Individual Prediction
1|Page
Topic # 23: Confidence Interval on 𝛽1and 𝛽2
Topic # 24: Testing of Hypothesis about the Regression Coefficient β
Topic # 25: ANOVA for Simple Linear Regression
Topic # 26: ANOVA for Multiple Linear Regressions
Topic # 27: Standard Error of Estimates
Topic # 28: Coefficient of Determination
Topic # 29: Coefficient of Multiple Determinations
Topic # 30: Dummy Variable
Topic # 31: Inverse Curve
Design of Experiment
Topic # 32: Latin – Square Design
Topic # 33: The Graeco – Latin Square Design
Topic # 34: Balance Incomplete Block Design
Topic # 35: Youden Square Design
Topic # 36: Completely Randomized Design
Topic # 37: Randomized Complete Block Design
Topic # 38: 𝟐𝟐 Factorial Experiment
Topic # 39: 𝟐𝟑 Factorial Experiment
Topic # 40: 𝟐𝟒 Factorial Experiment
Topic # 41: Blocking in Factorial Design
Topic # 42: Least Significant Difference (LSD)
Topic # 43: Duncan’s Test
Topic # 44: Tukey’s Test
Topic # 45: Regression of 𝟐𝟑 Factorial Design
Topic # 46: Dunnett’s Test
Topic # 47: Test of Equality of Variance
Topic # 48: Effect Size of One – Way (Eta-Square)
Quality Control
Topic # 49: Control Chart by the Cases in Units
Topic # 50: Control Chart by the Sub – Groups
Curvilinear
Regression
Topic # 51: Second Degree Curve Parabola
Topic # 52: Third Degree Curve Parabola
Topic # 53: Exponential Curve
Topic # 54: Power Curve
2|Page
Chi-Square Tests
Topic # 55: Testing of Hypothesis about the P’s of Multinomial – Distribution
Topic # 56: Goodness of Fit Test
Topic # 57: Testing of Hypothesis about the Equality of Several Proportions
Topic # 58: The Chi-Square Test as Test of Homogeneity
Time Series Analysis
Topic # 59: Fit the Second Degree Trend Curve
Topic # 60: Growth Model
Topic # 61: Autocorrelation Function (ACF)
Topic # 62: Partial Autocorrelation Correlation Faction (PACF)
Topic # 63: Time Series Graph
Topic # 64: Exponential Smoothing
Please Subscribe our Channel
for
Statistics and Economics
Learning and Solutions
https://www.youtube.com/channel/UCVOgn7b-ZGTUJUHWQ0I3miw/featured
3|Page
Books’ Key
Book # 1
Int to Stat-theory
Ex #
Pg #
Prb #
Author
= Introduction to Statistical Theory (Part – I)
= Example number
= Page number
= Problem number
= Sher Muhammad Chaudhry
Book # 2
Int to Stat-theory
Ex #
Pg #
Prb #
Author
= Introduction to Statistical Theory (Part – II)
= Example number
= Page number
= Problem number
= Sher Muhammad Chaudhry
Book # 3
Basic Eco
Sec #
Pg #
Author
= Basic Econometrics
= Section number
= Page number
= Damodar N. Gujrati
Book # 4
Stat for Eco & Mngmnt
Ex #
Pg #
Prb #
Author
= Statistics for Economics and Management
= Example number
= Page number
= Problem number
= Hamid A. Hakeem
Book # 5
Design & experiment
Ex #
Pg #
Author
= Design and analysis of experiment
= Example number
= Page number
= Douglas C. Montgomery
Book # 6
Int stat
Ex #
Sec #
Pg #
Author
= Introductory Statistics
= Example number
= Section number
= Page number
= Neil A. Weiss
4|Page
Book # 7
Busi Stat
Sec #
Pg #
Author
= Business Statistics
= Section number
= Page number
= Chris Robertson
Book # 8
Int to stat
Ex #
Sec #
Pg #
Author
= Introduction to Statistics
= Example number
= Section number
= Page number
= Ronald E. Walpole
Book # 9
Txt Busi Stat
Ex #
Pg #
Author
= Text Book of Business Statistics
= Example number
= Page number
= A. K. Sharma
Book # 10
Int linear Reg
Ex #
Pg #
Author
= Introduction To Linear Regression Analysis
= Example number
= Page number
= Douglas C. Montgomery
Book # 11
Statistics
Ex #
Pg #
Author
= Statistics
= Example number
= Page number
= Mc Clave. Dietrich. Sincich
Book # 12
Statistics bhvrl Scnc
Tbl #
Pg #
Author
= Statistics for the Behavioral Science
= Table number
= Page number
= Joan Welkowitz, Barry H. Cohen, Robert
5|Page
Topic # 1: Pie Chart
Represent the total expenditure and expenditures on various items of a family by a pie chart.
Items
Food
Clothing
House
rent
Fuel and
Light
Misc
Expenditure
50
30
20
15
35
(Int to Stat-theory, P-I) Ex # 2.13, Pg # 36
From the menus choose:
Analyze
Descriptive Statistics
Frequencies …
Clothing
Food
House Rent
Fuel & light
Miscellaneous
Topic # 2: Dot plots
A farmer interested in estimating his yield of oats if he farms organically. He uses the method on
a sample of 15 one – acre plots. The yields, in bushels, are depicted in following table. Construct
a dot plot for the data.
6|Page
67
Oats Yields
65 55 57
58
61
61
61
64
62
62
60
62
60
67
(Int Stat) Ex # 2.13, pg # 84
From the menus choose:
Graphs
Legacy Dialogs
Scatter/ Dot … (simple dot)
Oat Yields
55
58
60
62
65
68
Yield (bushels)
Topic # 3: Box plots
The capital assets ratio presented in the following table. Construct the Box plot.
3.6
5.4
3.6
3.2
2.7
US
UK
Ja
WG
Ja
3.7
3.1
2.8
5.8
4.4
US
WG
WG
SA
US
5.8
3.5
4.0
6.9
3.2
Fr
Ja
Ja
SA
It
1.4
3.6
5.3
3.3
3.2
Fr
US
UK
Ca
Ja
1.1
3.8
3.0
3.5
7.0
Fr
Ja
US
Nc
UK
1.6
3.6
0.4
2.8
3.6
Fr
Ja
Ja
WG
Ja
(Busi Stat) Sec # 3.4, pg # 47
From the menus choose:
Graphs
Legacy Dialogs
Boxplots … (simple)
7|Page
14
25
9
6.0
11
4.0
2.0
16
21
28
0.0
Capital_assest
Topic # 4: Cumulative Frequency Polygon
Construct the cumulative frequency polygon by using the following data.
Car Battery Lives
2.2
4.1
3.5
4.5
3.2
3.7
3
2.6
3.4
1.6
3.1
3.3
3.8
3.1
4.7
3.7
2.5
4.3
3.4
3.6
2.9
3.9
3.1
3.3
3.1
3.7
4.4
3.2
4.1
1.9
3.4
4.7
3.8
3.2
2.6
3.9
3
4.2
3.5
3.3
Make the frequency distribution at first of the ranges are, 1.5 – 1.9, 2.0 – 2.4, 2.5 – 2.9, 3.0 – 3.4,
3.5 – 3.9, 4.0 – 4.4 and 4.5 – 4.9.
(Int to stat) sec # 3.2, Pg # 56
From the menus choose:
Transform
Recode into Different Variables …
Graphs
Legacy Dialogs
Line … (simple)
8|Page
Cumulative Frequency
40
30
20
10
0
1.5 - 1.9
2.0 - 2.4
2.5 - 2.9
3.0 - 3.4
3.5 - 3.9
4.0 - 4.4
4.5 - 4.9
new
Topic # 5: Scatter Diagram
Construct the Scatter Diagram by using the following data, in which the X represents the age of
cars in the years, and Y represents the price of cars in the dollar ($).
Car
1
2
3
4
5
6
7
8
9
10
11
Age (Years)
5
4
6
5
5
5
6
6
2
7
7
Price (dollar $)
85
103
70
82
89
98
66
95
169
70
48
(Int Stat) Sec # 14.2, pg # 827
From the menus choose:
Graphs
Legacy Dialogs
Scatter/ Dot … (simple scatter)
9|Page
175
Price_Y
150
125
100
75
50
2
3
4
5
6
7
Age_X
Topic # 6: Scatter Plot Matrix
Draw a diagram of scatter plot matrix of 30 companies of from Fortune 500, November 1999.
Data set is given in the following table:
Company
General motors
Ford
Waal mart
Exxon
General electric
IBM
Citigourp
Phillip morris
Boeing
at & t
bank of America
Mobil
Hewlet packart
State frame ins cos
Sears roebuck
Ei du pont de numer
Protect and gambol
Tiaa-cref
Revenue
1.8
15.3
3.2
3.6
9.2
7.8
7.6
9.3
2.0
11.2
1.2
3.6
6.3
3.0
2.5
11.4
10.2
2.3
Asset
1.1
9.3
9.0
6.9
2.6
7.3
.9
9.0
3.0
10.7
.8
4.0
8.8
1.2
2.8
11.3
12.2
.3
Equity
19.7
94.3
21.0
14.6
23.9
32.6
13.6
33.2
9.1
25.1
11.2
9.3
17.4
3.2
17.3
31.5
30.9
13.3
Earning
4.18
17.76
1.98
2.55
2.80
6.57
2.43
2.20
1.15
3.55
2.90
2.10
2.77
.
2.68
3.90
2.56
.
Percentage
21.4
86.8
107.7
22.4
41.0
77.5
-6.8
22.8
-32.4
26.2
1.3
24.3
18.7
.
-4.4
-9.7
15.9
.
10 | P a g e
Merrill lynch
Prudential
Kmart
American international
chase Manhattan corp
Texaco
Bell atlantic
Funnie mae
Enron
Compaq
Morgan Stanley dean
Dayton Putson
3.5
3.2
1.7
11.3
11.7
1.8
9.4
10.8
2.2
-8.8
10.5
3.0
.4
.4
4.0
1.9
1.0
2.0
5.4
.7
2.4
-11.9
1.0
6.0
12.4
5.4
9.5
13.9
15.9
4.9
22.7
22.1
10.0
-24.2
23.2
17.6
3.00
.
1.01
3.57
4.27
.99
1.86
3.23
2.02
-1.71
5.33
1.98
-7.5
.
33.1
33.6
32.5
.5
22.5
31.7
39.9
49.0
21.4
62.1
(Busi Stat) Sec # 3.7.2, pg # 60
From the menus choose:
Graphs
Legacy Dialogs
Scatter/Dot … (matrix)
revenue
company
percentage
earning
equity
asset
american internation
at & t
bank of america
bell atilantic
boeing
chase manhattan corp
citigourp
compaq
dayton putson
ei du pont de numer
enron
exxon
ford
funnie mae
general electric
genral motors
hewlet packart
ibm
kmart
merrill lynch
mobil
morgan stanley dean
phillip morris
protect and gambol
sears roebuck
texaco
wal mart
revenue
asset
equity
earning
percentage
11 | P a g e
Topic # 7: Confidence Interval for µ
An Astronomer interesting in getting mean brightness of certain star got 6 reading on the stars
during his experiment. Calculate 99% confidence interval for the true value of the population
mean.
Reading
Brightness
1
30
2
32
3
29
4
31
5
33
6
30
(Stat for Eco & Mngmnt) Ex # 12.4, Pg # 308
From the menus choose:
Analyze
Descriptive Statistics
Explore …
Descriptives
X
Mean
99% Confidence
Interval for Mean
Statistic
30.83
28.41
Lower Bound
Upper Bound
Std. Error
.601
33.26
5% Trimmed Mean
30.81
Median
30.50
Variance
2.167
Std. Deviation
1.472
Minimum
29
Maximum
33
Range
4
Interquartilae Range
3
Skewness
Kurtosis
.418
.845
-.859
1.741
Topic # 8: Harmonic Mean
Compute the Harmonic Mean from the given data
Salary level
Hourly rate
No-of-Employee
1
45
25
2
50
20
3
55
15
4
60
10
5
65
5
(Stat for Eco & Mngmnt) Prb # 50, Pg # 70
12 | P a g e
From the menus choose:
Data
Weight cases …
Analyze
Compare Means
Means …
Report
Harmonic Mean
category
A
hourly rate
50.96
Total
50.96
Topic # 9: Trimmed Mean
Calculate the trimmed mean for the given data 45, 32, 37, 46, 39, 36, 41, 48 and 36.
(Int to Stat-theory, P-I) Ex # 4.12, Pg # 112
From the menus choose:
Analyze
Descriptive Statistics
Explore …
Descriptive
X
Statistic
40.00
Mean
95% Confidence
Interval for Mean
5% Trimmed Mean
Median
Variance
Std. Deviation
Lower Bound
Upper Bound
35.86
44.14
40.00
39.00
29.000
5.385
Minimum
32
Maximum
48
Range
16
Interquartile Range
Skewness
Kurtosis
Std. Error
1.795
10
.191
.717
-1.161
1.400
Topic # 10: Standard Scores (z Scores)
The distribution of the 20 base line heart rate scores expressed in beats per minute (bpm),
compute the Z-Scores on base line distribution.
13 | P a g e
Heart rate
(bpm)
72
70
70
70
69
69
68
68
68
68
Male
1
2
3
4
5
6
7
8
9
10
Male
11
12
13
14
15
16
17
18
19
20
Heat rate
(bpm)
67
67
67
67
67
67
66
66
66
64
(Statistics bhvral Scnc) Tbl # 7.1, Pg # 99
From the menus choose:
Analyze
Descriptive Statistics
Descriptives … (save stnd values)
Analyze
Reports
Case Summaries …
Z score of Heart Rate
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Total
Heart rate
72
70
70
70
69
69
68
68
68
68
67
67
67
67
67
67
66
66
66
64
72
Zscore(x)
2.30
1.21
1.21
1.21
.66
.66
.11
.11
.11
.11
-.44
-.44
-.44
-.44
-.44
-.44
-.99
-.99
-.99
-2.08
2.30
14 | P a g e
Topic # 11: Standard of Mean
A random sample of n = 6 has the elements 6, 10, 13, 14, 18 and 20 compute the standard error
of the mean.
(Int to Stat-theory, P-II) Ex # 15.1, Pg # 69
From the menus choose:
Analyze
Descriptive Statistics
Descrivtives …
Descriptive Statistics
N
Mean
Statistic
Y
6
Valid N (list wise)
6
Statistic
13.50
Std. Error
2.094
Topic # 12: Geometric Distribution
If the probability that a person will believes a rumor about the retirement of certain politician is
0.25, what is the probability that. The 6 people to hear the rumor will be the first to believe it.
(Int to Stat-theory, P-I) Ex # 8.26, Pg # 346
From the menus choose:
Transform
Compute Variable … (PDF)
Analyze
Reports
Case Summaries …
Geometric Distribution
1
x
6
probability
.059
Total
6
.059
Topic # 13: Exponential Distribution
The duration of the long distance telephone calls is found to be exponentially distributed with a
mean of 3 minutes. The mean of distribution is 3, so that λ = 1 / 3.What is the probability that a
call will last,
1. More than 3 minutes.
2. More than 5 minutes?
(Int to Stat-theory, P-I) Ex # 9.1, Pg # 365
15 | P a g e
Note: Here we use the cumulative distribution function (CDF) because, the probability is
required more than 3 minutes or 5 minutes. CDF calculate the probability less than the given
value therefore we subtract from total probability “1”.
From the menus choose:
Transform
Compute Variable … (CDF)
Analyze
Reports
Case Summaries …
Exponential distribution
1
x
3
Probability
.3679
2
5
.1889
Total
3
.3679
Topic # 14: Normal Distribution
Give a normal distribution with µ = 300 and σ = 50, find the probability that X assumes a value
greater than 362.
(Int to stat) Ex # 2, Pg # 189
Note: Here we use the cumulative distribution function (CDF) because, the probability is
required greater than 362. CDF calculate the probability less than the given value therefore we
subtract from total probability “1”.
From the menus choose:
Transform
Compute Variable … (CDF)
Analyze
Reports
Case Summaries …
Normal Distribution
1
x
362
probability
.1075
Total
362
.1075
16 | P a g e
Topic # 15: Uniform Distribution
An unprincipled used car dealer sells a car to an unsuspected buyer, even thought the dealer
knows that the car will have a major breakdown within the next 6 months. Assume that x is a
uniform random variable with values between 0 and 6 months. We must find the probability that
between the points x = 0 and x = 1.5. P (0 < x < 1.5) =?
(Statistics) Ex # 5.1, Pg # 196
From the menus choose:
Transform
Compute Variable … (CDF)
Analyze
Reports
Case Summaries …
UNIFORM DISTRIBUTION
1
X
1.5
probability
.25
Total
1.5
.25
Topic # 16: Correlation for Grouped Data
Compute correlation coefficient from the following correlation table for weights and heights of
women students.
Heights
in inches
57
60
63
66
69
72
90
--8
3
1
-----
110
--21
50
24
1
---
Weights in pounds
130
150
----8
1
57
12
54
19
8
5
2
---
170
1
----3
3
---
190
----2
----1
(Int to Stat-theory, P-I) Prb # 10.26, Pg # 455
From the menus choose:
Data
Weight cases …
Analyze
Descriptive Statistics
Crosstabs …
17 | P a g e
Symmetric Measures
Asymp.
Std.
Error(a)
Value
.388
Interval by Interval
Pearson's R
Ordinal by Ordinal
Spearman Correlation
.395
N of Valid Cases
Approx.
T(b)
Approx. Sig.
.065
7.068
.000(c)
.053
7.213
.000(c)
284
Topic # 17: Rank Correlation
Find the coefficient of rank correlation from the following ranking of six values between two
data sets?
A
B
7.4
8.5
9.0
6.1
11.0
2.4
2.5
6.7
4.6
12.6
6.5
3.3
(Int to Stat-theory, P-I) Prb # 10.30, Pg # 455
From the menus choose:
Analyze
Correlate
Bivariate …
Correlations
A
Spearman's rho
A
Correlation Coefficient
Sig. (2-tailed)
N
B
Correlation Coefficient
Sig. (2-tailed)
N
1.000
B
-.600
.
.208
6
-.600
1.000
6
.208
.
6
6
Topic # 18: Rank Correlation for Tied Ranks
Two members of selection committee rank eight persons according to their suitably for
promotion as follows.
Persons
Member – 1
Member – 2
A
1
2
B
2.5
4
C
2.5
1
D
4
3
E
5
6
F
6
6
G
7
6
H
8
8
Calculate the coefficient rank correlation?
(Int to Stat-theory, P-I) Ex # 10.11, Pg # 446
18 | P a g e
From the menus choose:
Analyze
Correlate
Bivariate …
Correlations
member1
Spearman's rho
member1
Correlation Coefficient
1.000
member2
.896
Sig. (2-tailed)
.
.003
8
Correlation Coefficient
8
.896
1.000
Sig. (2-tailed)
.003
.
8
8
N
member2
N
Topic # 19: Covariance
Find out the covariance between the following variables X and Y are shown in the table,
Country
Japan
Italy
UK
W. Germany
USA
Spain
Australia
New Zealand
X
0.28
0.33
0.41
0.41
0.43
0.44
0.49
0.71
Y
3.7
2.9
2.2
2.4
2.1
1.6
1.8
0.3
(Busi Stat) Sec # 6.4.1, pg # 128
From the menus choose:
Analyze
Scale
Reliability Analysis …
(covariance)
Inter-Item Covariance Matrix
X
x
y
.016
-.122
y
-.122
.982
19 | P a g e
Topic # 20: Test Of linearity Of Regression
The following data show the heights (X) and weight (Y) of twelve men. We selected the height in
advance and then observed the weights of a random group of men having the selected heights.
X (height)
Y (weight)
60
60
60
62
62
62
62
64
64
70
70
70
110 135 120 120 140 130 135 150 145 170 185 160
Test the hypothesis at the 0.05 level of significance, that the regression is linear.
(Int to Stat-theory, P-II) Ex # 21.7, Pg # 366
From the menus choose:
Analyze
Compare Means
Means …
ANOVA Table
Y*
X
Between
Groups
Sum of
Squares
4402.08
3
4341.81
2
(Combined)
Linearity
Deviation from
Linearity
Within Groups
Total
Mean
Square
df
F
Sig.
3
1467.361
13.578
.002
1
4341.812
40.175
.000
60.271
2
30.136
.279
.764
864.583
5266.66
7
8
108.073
11
Topic # 21: Mean Prediction
On the basis of the sample data of 3.2 we obtained the following sample regression 𝑌𝑖 = 24.4545
+ 0.5091𝑋𝑖 . Assume that 𝑋0 = 100 and we want to predict 95 % confidence interval for Mean
prediction and Individual prediction.
X ($)
Y ($)
80
70
100
65
120
90
140
95
160
110
180
115
200
120
220
140
240
155
260
150
(Basic Eco) Sec # 5.10, Pg # 137
From the menus choose:
Analyze
Regression
Linear …
Analyze
Reports
Case Summaries …
20 | P a g e
Mean prediction
2
80
100
70
65
95% L CI
for Y mean
56.38144
67.89991
3
120
90
79.26830
91.82261
4
140
95
90.38565
101.06890
5
160
110
101.10304
110.71514
6
180
115
111.28486
120.89696
7
200
120
120.93110
131.61435
8
220
140
130.17739
142.73170
9
240
155
139.17264
154.10009
10
260
150
148.01781
165.61856
80
70
56.38144
73.98219
X
1
Y
Total
95% U CI
for Y mean
73.98219
82.82736
Topic # 22: Individual Prediction
On the basis of the sample data of 3.2 we obtained the following sample regression 𝑌𝑖 = 24.4545
+ 0.5091𝑋𝑖 . Assume that 𝑋0 = 100 and we want to predict 95 % confidence interval for Mean
prediction and Individual prediction.
X ($)
Y ($)
80
70
100
65
120
90
140
95
160
110
180
115
200
120
220
140
240
155
260
150
(Basic Eco) Sec # 5.10, Pg # 137
From the menus choose:
Analyze
Regression
Linear …
Analyze
Reports
Case Summaries …
Individual Prediction
X
1
2
3
4
5
6
7
8
9
10
Total
N
Y
80
100
120
140
160
180
200
220
240
260
10
70
65
90
95
110
115
120
140
155
150
10
95% L CI for Y
individual
47.81420
58.63358
69.30999
79.83009
90.18377
100.36559
110.37554
120.21909
129.90631
139.45057
10
95% U CI for
Y individual
82.54943
92.09369
101.78091
111.62446
121.63441
131.81623
142.16991
152.69001
163.36642
174.18580
10
21 | P a g e
Topic # 23: Confidence Intervals for Regression Coefficients β1 and β2
Construct the 95% confidence intervals on the regression coefficients β1 and β2 for the given
data.
X ($)
Y ($)
80
70
100
65
120
90
140
95
160
110
180
115
200
120
220
140
240
155
260
150
(Basic Eco) Sec # 5.3, Pg # 117
From the menus choose:
Analyze
Regression
Linear …
Coefficients
95% Confidence Interval for B
Model
1
Lower Bound
9.664
.427
(Constant)
X
Upper Bound
39.245
.592
Topic # 24: Testing of Hypothesis for Regression Coefficient β
Estimate the regression line from the following data of height (X) and weight (Y) of 12 persons.
Heights (X)
60
62
64
70
Weights (Y)
110,135,120
120,140,130,135
150,145
170,185,160
Test the hypothesis that the population regression coefficient β = 0. Use 0.05 level of
significance.
(Int to Stat-theory, P-II) Ex # 21.5, Pg # 359
From the menus choose:
Analyze
Regression
Linear …
Coefficients
Unstandardized
Coefficients
Model
1
(Constant)
X
B
-179.359
5.029
Std. Error
46.936
.734
Standardized
Coefficients
Beta
.908
t
B
-3.821
6.852
Sig.
Std. Error
.003
.000
22 | P a g e
Topic # 25: ANOVA for Simple Linear Regression and Test of Ho: β = 0
Test the hypothesis that β = 0 at the 0.05 level of significance for the given data by setting the
results in an ANOVA table.
Heights (X)
60
62
64
70
Weights (Y)
110,135,120
120,140,130,135
150,145
170,185,160
(Int to Stat-theory, P-II) Ex # 21.5, Pg # 359
From the menus choose:
Analyze
Regression
Linear …
ANOVA
Model
1
Sum of
Squares
df
Mean Square
Regression
4341.812
1
4341.812
Residual
924.854
5266.667
10
11
92.485
Total
F
46.946
Sig.
.000(a)
Conclusion:
The value of the F statistic is 46.946 that fall in critical region, and the probability of
significance (Sig) is 0.00, therefore we reject our null hypothesis and conclude that regression
coefficient for simple linear regression is not zero, it has some specific slope.
Topic # 26: ANOVA for Multiple Regression and Testing Hypothesis about the β Parameter
Consider the following set of data:
Y
X1
X2
4
1
2
6
1
4
5
2
3
8
3
4
12
4
8
13
5
8
19
7
11
Obtain the ANOVA table and test the hypothesis that there is no association between both
regressor and the dependent variables. Use the 0.01 level of significance.
(Int to Stat-theory, P-II) Ex # 21.16, Pg # 386
23 | P a g e
From the menus choose:
Analyze
Regression
Linear …
ANOVA
Model
1
Sum of
Squares
172.053
Regression
Residual
df
2
1.661
173.714
Total
Mean Square
86.027
4
6
F
207.154
Sig.
.000(a)
.415
Conclusion:
The value of the F statistic is 207.154 that fall in critical region, and the probability of
significance (Sig) is 0.00, therefore we reject our null hypothesis and conclude that regression
coefficient for multiple linear regression is not zero, it has some specific slope.
Topic # 27: Standard Error of Estimate
Using the following data, find the standard error of estimate 𝑆𝑦.𝑥 ?
X
Y
5
16
6
19
8
23
10
28
12
36
13
41
15
44
16
45
17
50
(Int to Stat-theory, P-I) Ex # 10.3, Pg # 431
From the menus choose:
Analyze
Regression
Linear …
Model Summary
Model
1
R
.994(a)
R Square
.987
Adjusted R
Square
.985
Std. Error of
the Estimate
1.507
Topic # 28: Coefficient of Determination
Using the following data, find the Coefficient of determination?
X
Y
5
85
4
103
6
70
5
82
5
89
5
98
6
66
6
95
2
169
7
70
7
48
(Int Stat) Ex # 14.7, pg # 845
24 | P a g e
From the menus choose:
Analyze
Regression
Linear …
Model Summary
Model
1
R
.924(a)
R Square
.853
Adjusted R
Square
.837
Std. Error of
the Estimate
12.577
Topic # 29: Coefficient of Multiple Determinations
A statistician wants to predict the incomes of restaurant, using two independent variables the
number of restaurant employees and restaurant floor area. He calculated the following data:
Income
(000) Y
30
22
16
7
14
Floor area
(000 sq. ft)
Number of
employees
10
5
10
3
2
15
8
12
7
10
𝑋1
𝑋2
Compute the standard error of estimate, coefficient of multiple determinations?
(Int to Stat-theory, P-I) Ex # 11.3, Pg # 463
From the menus choose:
Analyze
Regression
Linear …
Model Summary
Model
1
R
.738(a)
R Square
.545
Adjusted R
Square
.089
Std. Error of
the Estimate
8.276
Topic # 30: Dummy Variable
Twenty observations of tool life and lathe speed are presented in table. Where the hours are a
dependent variable, use the dummy variable Tool type and estimate the regression equation.
25 | P a g e
Speed_arpm
(𝑋1 )
610
950
720
840
980
530
680
540
890
730
670
770
880
1000
760
590
910
650
810
500
Hours (Y)
18.73
14.52
17.43
14.54
13.44
24.39
13.34
22.71
12.68
19.32
30.16
27.09
25.40
26.05
33.49
35.62
26.07
36.78
34.95
43.67
Where the
Tool Type (𝑋2 )
A
A
A
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
B
B
0 𝑖𝑓 𝑡ℎ𝑒 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛 𝑖𝑠 𝑓𝑟𝑜𝑚 𝑡𝑜𝑜𝑙 𝑡𝑦𝑝𝑒 𝐴
𝑋2 = {
1 𝑖𝑓 𝑡ℎ𝑒 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛 𝑖𝑠 𝑓𝑟𝑜𝑚 𝑡𝑜𝑜𝑙 𝑡𝑦𝑝𝑒 𝐵
(Int linear Reg) Ex # 8.1, Pg # 267
From the menus choose:
Analyze
Regression
Linear …
Note: before analyze the data, first we values labeled A as 0 and B as 1.
Coefficients
Unstandardized
Coefficients
Model
1
(Constant)
x
tool
B
36.986
Std. Error
Standardized
Coefficients
t
Sig.
Beta
B
Std. Error
10.536
.000
-.027
3.510
.005
-.451
-5.887
.000
15.004
1.360
.845
11.035
.000
26 | P a g e
Topic # 31: Inverse Curve
Fit the inverse curve on the following data,
Year
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
Unemployment
1.4
1.1
1.5
1.5
1.2
1.0
1.1
1.3
1.8
1.9
1.5
1.4
1.8
2.1
1.5
1.3
1.4
Wage rate
1.8
8.5
8.4
4.5
4.3
6.9
8.0
5.0
3.6
2.6
2.6
4.2
3.6
3.7
4.8
4.3
4.6
(Basic Eco) Sec # 6.7, Pg # 177
From the menus choose:
Analyze
Regression
Curve Estimation …
Model Summary and Parameter Estimates
Model Summary
Equation
Inverse
R Square
.385
F
9.385
df1
Parameter Estimates
df2
1
Sig.
15
.008
Constant
-1.428
b1
8.724
27 | P a g e
Topic # 32: Latin Square Design
Five fertilizers A, B, C, D and E were tested by the arranging plants in a Latin Square design in a
field. The rows and columns in the table are rows and columns in the field. The yields per plot
are as shown
COLUMNS
1
2
3
4
5
1
B 4.9
D 6.4
E 3.3
A 9.5
C 11.8
2
C 9.3
A 4.0
B 6.2
E 5.1
D 5.4
ROWS3
D 7.6
C 15.4
A 6.5
B 6.0
E 4.6
4
E 5.3
B 7.6
C 13.2
D 8.6
A 4.9
5
A 9.3
E 6.3
D 11.8
C 15.9
B 7.6
Analyze the data for evidence at 5% level that the mean yields are not equal for the five
fertilizers.
(Int to Stat-theory, P-II) Ex # 23.6, Pg # 446
From the menus choose:
Analyze
General Linear Model
Univariate …
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
Rows
Columns
Type III Sum
of Squares
df
Mean Square
21.441
9.175
.000
1544.490
1
1544.490
660.886
.000
46.668
4
11.667
4.992
.013
1.500
21.032
.263
14.020
4
4
3.505
49.152
Error
28.044
12
2.337
Total
1829.830
285.340
25
24
Corrected Total
Sig.
12
196.608
treatments
F
257.296(a)
.000
Conclusion:
Since the computed value of F = 21. 03 falls in the critical region, we therefore reject our
null hypothesis. Hence we conclude that the data provide sufficient evidence to indicate at the
5% significance level that the mean yields for the fertilizers are not equal.
28 | P a g e
Multiple Comparisons
LSD
Mean
Difference
(I-J)
.380
-6.280(*)
Std. Error
.9669
.9669
Sig.
.701
.000
Lower bound
-1.727
-8.387
Upper bound
2.487
-4.173
D
-1.120
.9669
.269
-3.227
.987
E
1.920
.9669
.070
-.187
4.027
A
-.380
.9669
.701
-2.487
1.727
C
D
E
-6.660(*)
-1.500
1.540
.9669
.9669
.9669
.000
.147
.137
-8.767
-3.607
-.567
-4.553
.607
3.647
A
6.280(*)
.9669
.000
4.173
8.387
B
6.660(*)
.9669
.000
4.553
8.767
D
5.160(*)
.9669
.000
3.053
7.267
E
A
B
8.200(*)
1.120
1.500
.9669
.9669
.9669
.000
.269
.147
6.093
-.987
-.607
10.307
3.227
3.607
C
-5.160(*)
.9669
.000
-7.267
-3.053
E
3.040(*)
.9669
.008
.933
5.147
A
-1.920
.9669
.070
-4.027
.187
-1.540
.9669
-8.200(*)
.9669
-3.040(*)
.9669
* The mean difference is significant at the .05 level.
.137
.000
.008
-3.647
-10.307
-5.147
.567
-6.093
-.933
(I)
treatments
A
B
C
D
E
(J)
treatments
B
C
95% Confidence Interval
B
C
D
Topic # 33: The Graeco–Latin Square Design
Suppose that in the dynamite formulation experiment of example an additional factor, test
assemblies, could be of importance. Let there be five test assemblies denoted by the Greek letters
α (alpha), β (beta), γ (gamma), δ (delta) and ε (epsilon). The resulting 5×5 Graeco – Latin
Square design is shown in following table.
Operators
1
2
3
4
5
1
A α -1
B γ -5
C ε -6
D β -1
E δ -1
2
B β -8
C δ -1
Dα 5
Eγ 2
A ε 11
Raw Material 3
C γ -7
D ε 13
Eβ 1
Aδ 2
B α -4
4
Dδ 1
Eα 6
Aγ 1
B ε -2
C β -3
5
E ε -3
Aβ 5
B δ -5
Cα 4
Dγ 6
(Design & experiment) Ex # 4.4, Pg # 152
29 | P a g e
From the menus choose:
Analyze
General Linear Model
Univariate …
Tests of Between-Subjects Effects
Source
Corrected
Model
Intercept
Type III Sum
of Squares
df
Mean Square
610.000(a)
16
38.125
F
Sig.
4.621
.017
4.000
1
4.000
.485
.506
Raw Material
68.000
4
17.000
2.061
.178
Operators
Formulation
150.000
4
37.500
4
82.500
4.545
10.000
.033
330.000
62.000
4
15.500
1.879
.208
8.250
Greek
Error
66.000
8
Total
680.000
676.000
25
24
Corrected Total
.003
Conclusion:
Since the computed value of F = 10 falls in the critical region, we therefore reject our null
hypothesis. Hence we conclude that the data provide sufficient evidence to indicate at the 5%
significance level that the mean yields for the Formulation are not equal.
Multiple Comparisons
LSD
(I) Formulation
A
(J) Formulation
B
C
D
B
C
Mean
Difference
(I-J)
8.40(*)
6.20(*)
Std. Error
1.817
1.817
Sig.
.002
.009
Lower bound
4.21
2.01
Upper bound
12.59
10.39
-5.39
2.99
95% Confidence Interval
-1.20
1.817
.527
E
2.60
1.817
.190
-1.59
6.79
A
-8.40(*)
1.817
.002
-12.59
-4.21
C
D
E
-2.20
-9.60(*)
-5.80(*)
1.817
1.817
1.817
.260
.001
.013
-6.39
-13.79
-9.99
1.99
-5.41
-1.61
A
-10.39
-2.01
-6.20(*)
1.817
.009
B
2.20
1.817
.260
-1.99
6.39
D
-7.40(*)
1.817
.004
-11.59
-3.21
E
A
B
-3.60
1.20
9.60(*)
1.817
1.817
1.817
.083
.527
.001
-7.79
-2.99
5.41
.59
5.39
13.79
C
7.40(*)
1.817
.004
3.21
11.59
E
3.80
1.817
.070
-.39
7.99
A
-2.60
1.817
.190
-6.79
1.59
5.80(*)
C
3.60
D
-3.80
* The mean difference is significant at the .05 level.
1.817
1.817
1.817
.013
.083
.070
1.61
-.59
-7.99
9.99
7.79
.39
D
E
B
30 | P a g e
Topic # 34: Balance Incomplete Block Design
Consider the data in the given table for the Catalyst experiment. This is a balanced incomplete
block design with a = 4, b = 4, k = 3, r = 3, λ = 2 and N = 12. Construct the ANOVA table.
Block (Raw Material)
2
3
Treatment
(Catalyst)
1
1
73
74
---
71
2
---
75
67
72
3
73
75
68
---
4
75
---
72
75
4
(Design & experiment) Ex # 4.5, Pg # 156
From the menus choose:
Analyze
General Linear Model
Univariate …
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
Type III Sum
of Squares
df
Mean Square
F
Sig.
77.750(a)
6
12.958
19.936
.002
63075.000
1
63075.000
97038.462
.000
treatments
22.750
3
7.583
11.667
.011
Blocks
33.889
.001
66.083
3
22.028
Error
3.250
5
.650
Total
63156.000
12
81.000
11
Corrected Total
Conclusion:
The value of the F statistic is 11.667 that fall in critical region, and the probability of
significance (Sig) is 0.01, therefore we reject our null hypothesis and conclude that the all
treatment’s mean are not equal.
31 | P a g e
Multiple Comparisons
LSD
Mean
Difference
(I-J)
1.33
.67
Std. Error
.658
.658
Sig.
.099
.358
Lower bound
-.36
-1.03
Upper bound
3.03
2.36
4
-1.33
.658
.099
-3.03
.36
1
-1.33
.658
.099
-3.03
.36
3
4
1
-.67
-2.67(*)
-.67
.658
.658
.658
.358
.010
.358
-2.36
-4.36
-2.36
1.03
-.97
1.03
2
.67
.658
.358
-1.03
2.36
4
-2.00(*)
.658
.029
-3.69
-.31
1.33
.658
2
2.67(*)
.658
3
2.00(*)
.658
* The mean difference is significant at the .05 level.
.099
.010
.029
-.36
.97
.31
3.03
4.36
3.69
(I)
treatments
1
2
3
4
(J)
treatments
2
3
95% Confidence Interval
1
Topic # 35: Youden Squares Design
An industrial engineering is studying the effect of five illumination levels on the occurrence of
defects in an assembly operation. Because time may be factor in the experiment, she has decided
to run the experiment in five blocks, where each block is a day of the week. However, the
department in which the experiment is conducted has four work stations, and these stations
represent a potential source of variability. The engineer decided to run a Youden Square with
five rows (block_days), four columns (work_station), and five treatments (illumination_level).
Work Station
2
3
Day
(Block)
1
1
A=3
B=1
C = -2
D=0
2
B=0
C=0
D = -1
E=7
3
C = -1
D=0
E=5
A=3
4
D = -1
E=6
A=4
B=0
5
E=5
A=2
B=1
C = -1
4
(Design & experiment) Ex # 6.2, Pg # 181
From the menus choose:
Analyze
General Linear Model
Univariate …
32 | P a g e
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
Type III Sum
of Squares
Illumination_level
Mean Square
F
Sig.
128.417(a)
11
11.674
14.295
.000
48.050
1
48.050
58.837
.000
.867
4
.217
.265
.892
1.350
3
.450
.551
.662
120.367
4
30.092
36.847
.000
.817
Block_days
Work_station
df
Error
6.533
8
Total
183.000
20
Corrected Total
134.950
19
Conclusion:
The value of the F statistic is 36.847 that fall in critical region, and the probability of
significance (Sig) is 0.00, therefore we reject our null hypothesis and conclude that the all
illumination levels are not equal.
Multiple Comparisons
LSD
(I)
Illumination_level
A
B
C
D
E
95% Confidence Interval
Lower
Upper
bound
bound
1.03
3.97
2.53
5.47
Mean
Difference
(I-J)
2.50(*)
4.00(*)
Std.
Error
.639
.639
Sig.
.004
.000
D
3.50(*)
.639
.001
2.03
4.97
E
-2.75(*)
.639
.003
-4.22
-1.28
A
-2.50(*)
.639
.004
-3.97
-1.03
C
D
E
1.50(*)
1.00
-5.25(*)
.639
.639
.639
.047
.156
.000
.03
-.47
-6.72
2.97
2.47
-3.78
A
-4.00(*)
.639
.000
-5.47
-2.53
B
-1.50(*)
.639
.047
-2.97
-.03
D
-.50
.639
.456
-1.97
.97
E
A
B
-6.75(*)
-3.50(*)
-1.00
.639
.639
.639
.000
.001
.156
-8.22
-4.97
-2.47
-5.28
-2.03
.47
C
.50
.639
.456
-.97
1.97
E
-6.25(*)
.639
.000
-7.72
-4.78
A
2.75(*)
.639
.003
1.28
4.22
.639
.639
.639
.000
.000
.000
3.78
5.28
4.78
6.72
8.22
7.72
(J)
Illumination_level
B
C
B
5.25(*)
C
6.75(*)
D
6.25(*)
* The mean difference is significant at the .05 level.
33 | P a g e
Topic # 36: The Complete Randomized Design.
An experiment was conducted to compare the yields of varieties of potatoes. Each variety was
assigned at random to equal size plot, four times. The yield were as follows,
Variety
A
23
26
20
17
B
18
28
17
21
C
16
25
12
14
Test the hypothesis that the three varieties of potatoes are not different in the yield capabilities.
(Int to Stat-theory, P-II) Ex # 23.1, Pg # 424
From the menus choose:
Analyze
Compare Means
One – Way ANOVA …
ANOVA
Sum of
Squares
Between Groups
df
Mean Square
54.500
2
27.250
Within Groups
217.750
9
24.194
Total
272.250
11
F
1.126
Sig.
.366
Conclusion:
Since the computed value of F = 1.13 does not fall in the critical region, so we accept our
null hypothesis and may conclude that there is no difference among the yielding capabilities of
the three varieties of potatoes.
Topic # 37: The Randomized Complete Block Design.
Four varieties of wheat were tried in a randomized complete block design in four replications.
Yields in KG per plot are shown in the table given below. Test of hypothesis that there are no
differences in the means of four varieties.
Replicates
I
II
III
IV
V1
2
2
4
1
Varieties
V2
V3
5
4
3
3
6
6
4
2
V4
1
1
2
3
(Int to Stat-theory, P-II) Ex # 23.2, Pg # 429
34 | P a g e
From the menus choose:
Analyze
General linear Model
Univariate…
Tests of Between-Subjects Effects
Source
Corrected Model
Type III Sum
of Squares
df
Mean Square
F
Sig.
31.875(a)
6
5.313
5.276
.014
150.063
1
150.063
.000
Replicates
12.188
3
4.063
149.028
4.034
Varieties
19.687
3
6.562
6.517
.012
1.007
Intercept
Error
9.063
9
Total
191.000
16
40.938
15
Corrected Total
.045
Conclusion:
Since the computed value of F for the replicates and for the Varieties both falls in the
critical region, we therefore reject our null hypotheses and conclude that the means of the
Replications and means of the four varieties of wheat are significantly different.
Furthermore, since F-statistic indicates rejection of the null hypothesis, we therefore
apply the LSD on varieties to find out which means differ significantly.
Multiple Comparisons
LSD
Mean
Difference
(I-J)
-2.25(*)
-1.50
Std. Error
.710
.710
Sig.
.011
.064
Lower bound
-3.86
-3.11
Upper bound
-.64
.11
4
.50
.710
.499
-1.11
2.11
1
2.25(*)
.710
.011
.64
3.86
3
.75
2.75(*)
1.50
.710
.710
.710
.318
.004
.064
-.86
1.14
-.11
2.36
4.36
3.11
2
-.75
.710
.318
-2.36
.86
4
2.00(*)
.710
.020
.39
3.61
-.50
.710
-2.75(*)
.710
-2.00(*)
.710
* The mean difference is significant at the .05 level.
.499
.004
.020
-2.11
-4.36
-3.61
1.11
-1.14
-.39
(I) varieties
1
2
3
4
(J) varieties
2
3
4
1
1
2
3
95% Confidence Interval
35 | P a g e
Topic # 38: 𝟐𝟐 Factorial Experiment
The data of the following table are from a 2 × 2 factorial experiment. Partition the treatment sum
of squares into main effects and interaction component.
Treatment combinations
A0b0
6
14
8
9
7
A0b1
12
14
13
11
13
a1b0
26
17
21
30
27
a1b1
21
16
20
17
21
(Int to Stat-theory, P-II) Prb # 23.37, Pg # 474
From the menus choose:
Analyze
General Linear Model
Univariate … (full factorial)
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
A
B
Type III Sum
of Squares
df
Mean Square
Sig.
3
232.583
21.535
.000
5216.450
1
5216.450
.000
594.050
1
594.050
483.005
55.005
2.450
1
2.450
.227
.640
101.250
10.800
9.375
.007
A*B
101.250
1
Error
172.800
16
Total
6087.000
870.550
20
19
Corrected Total
F
697.750(a)
.000
Conclusion:
Since the computed values of F for the treatment combinations is 21.535 regarding
treatment combinations falls in critical region., so we reject null hypothesis, that there is no
significant difference between combinations. As null hypothesis is rejected, we therefore test the
significance of differences between treatment A and treatment B, etc.
We reject hypothesis about treatment A’ effects are equal.
We accept hypothesis about treatment B’ effects are equal.
We reject hypothesis about there in no interaction between treatment A and B.
36 | P a g e
Topic # 39: 𝟐𝟑 Factorial Experiment
The following data were obtained from a 𝟐𝟑 factorial experiment repeated 3 times
Replication
1
2
3
(1)
A
Treatment combinations
b ab
C
ac
bc
12
19
10
15
20
16
24
16
17
23
17
27
17
25
21
16
19
19
24
23
29
abc
28
25
20
(Int to Stat-theory, P-II) Prb # 23.41, Pg # 476
From the menus choose:
Analyze
General Linear Model
Univariate …
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
A
Type III Sum
of Squares
323.167(a)
df
7
9680.167
1
Mean Square
46.167
F
3.095
Sig.
.029
648.950
.000
2.667
1
9680.167
2.667
.179
.678
B
170.667
1
170.667
11.441
.004
C
104.167
1
104.167
6.983
.018
A*B
1.500
1
1.500
.101
.755
A*C
42.667
1
42.667
2.860
.110
B*C
.000
1
.000
.000
1.000
1.500
1
1.500
.101
.755
Error
238.667
16
14.917
Total
10242.000
24
561.833
23
A*B*C
Corrected Total
Topic # 40: 24 Factorial Experiment
A chemical product is produced in a pressure vessel. A factorial experiment is carried out in the
plot plant to study the factor thought to influence the filtration rate of this product. The four
factors are temperature (A), pressure (B), concentration of formaldehyde (C), and stirring rate
(D). Perform 24 factorial experiment for following data.
37 | P a g e
Run Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Run Label
(1)
A
B
ab
C
ac
bc
abc
D
ad
bd
abd
cd
acd
bcd
abcd
Filtration Rate
45
71
48
65
68
60
80
65
43
100
45
104
75
86
70
96
(Design & experiment) Ex # 6.2, Pg # 246
From the menus choose:
Analyze
General Linear Model
Univariate …
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
A
B
C
D
A*B
A*C
B*C
A*B*C
A*D
B*D
A*B*D
C*D
A*C*D
B*C*D
A*B*C*D
Error
Total
Corrected Total
Type III Sum
of Squares
5730.937(a)
78540.063
1870.563
39.063
390.063
855.563
.063
1314.063
22.563
14.063
1105.563
.563
68.063
5.063
10.563
27.563
7.563
.000
84271.000
5730.937
df
Mean Square
15
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
16
15
382.062
78540.063
1870.563
39.063
390.063
855.563
.063
1314.063
22.563
14.063
1105.563
.563
68.063
5.063
10.563
27.563
7.563
.
38 | P a g e
Topic # 41: Blocking in Factorial Design
An engineer is studying methods for improving the ability to detect targets on a radar scope. Two
factors she considers to be important are the amount of background noise, or “Ground Clutter”
on the scope and “The Type of Filter” placed over the screen.
Operator(blocks)
Filter Types
Ground Clutter
Low
Medium
High
1
2
3
4
1
2
1
2
1
2
1
2
90
102
114
86
87
93
96
106
112
84
90
91
100
105
108
92
97
95
92
96
98
81
80
83
Perform the analysis and make ANOVA table.
(Design & experiment) Ex # 5.6, Pg # 207
From the menus choose:
Analyze
General Linear Model
Univariate … (interaction)
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
Ground clutter
Blocks
filter types
Ground clutter *
filter types
Error
Total
Corrected Total
Type III Sum
of Squares
1881.500(a)
df
8
Mean Square
235.188
F
21.209
Sig.
.000
216220.167
1
216220.167
2
167.792
19498.813
15.132
.000
335.583
.000
402.167
3
134.056
1
1066.667
12.089
96.192
.000
1066.667
77.083
2
38.542
3.476
.058
11.089
166.333
15
218268.000
24
2047.833
23
.000
Conclusion:
Since the computed value of F for the Ground clutter and for the Filter types both falls in
the critical region, we therefore reject our null hypotheses and conclude that the means of the
Ground clutter and means of the Filter types are significantly different. But it is noted that there
is no interaction between the Ground clutter and Filter types.
Furthermore, since F-statistic indicates rejection of the null hypothesis, we therefore
apply the LSD on Ground clutter to find out which means differ significantly.
39 | P a g e
Multiple Comparisons
LSD
(I)
Ground_clutter
low
(J)
Ground_clutter
medium
high
medium
low
95% Confidence Interval
Lower
Upper
bound
bound
-8.80
-1.70
-12.67
-5.58
Mean
Difference
(I-J)
-5.25(*)
-9.13(*)
Std.
Error
1.665
1.665
Sig.
.007
.000
5.25(*)
1.665
.007
1.70
8.80
1.665
1.665
1.665
.034
.000
.034
-7.42
5.58
.33
-.33
12.67
7.42
high
-3.88(*)
high
low
9.13(*)
medium
3.88(*)
* The mean difference is significant at the .05 level.
Topic # 42: Least Significant Difference (L.S.D)
Perform the analysis of variance on the following data and analyze the treatment means using the
LSD test with a 0.05 level of significance.
Treatments
1
2
Blocks3
4
1
2
3
4
5
6
1
1
3
2
3
4
6
3
6
4
7
2
4
8
8
3
3
5
4
2
2
1
3
1
(Int to Stat-theory, P-II) Ex # 20.7, Pg # 325
From the menus choose:
Analyze
General Linear Model
Univariate …
40 | P a g e
Multiple Comparisons
LSD
(I)
Treatment
1
2
3
4
5
95% Confidence Interval
Lowe
Upper
bound
bound
-4.14
-.36
-4.89
-1.11
Mean
Difference
(I-J)
-2.25(*)
-3.00(*)
Std.
Error
.885
.885
Sig.
.023
.004
4
-4.00(*)
.885
.000
-5.89
5
-1.75
.885
.067
-3.64
.14
6
.00
.885
1.000
-1.89
1.89
1
2.25(*)
.885
.023
.36
4.14
3
4
5
-.75
-1.75
.50
.885
.885
.885
.410
.067
.580
-2.64
-3.64
-1.39
1.14
.14
2.39
6
2.25(*)
.885
.023
.36
4.14
1
(J)
Treatment
2
3
-2.11
3.00(*)
.885
.004
1.11
4.89
2
.75
.885
.410
-1.14
2.64
4
-1.00
.885
.276
-2.89
.89
5
6
1
1.25
3.00(*)
4.00(*)
.885
.885
.885
.178
.004
.000
-.64
1.11
2.11
3.14
4.89
5.89
2
1.75
.885
.067
-.14
3.64
3
1.00
.885
.276
-.89
2.89
5
2.25(*)
.885
.023
.36
4.14
6
4.00(*)
.885
.000
2.11
5.89
1
2
3
1.75
-.50
-1.25
.885
.885
.885
.067
.580
.178
-.14
-2.39
-3.14
3.64
1.39
.64
4
-2.25(*)
.885
.023
-4.14
-.36
6
1.75
.885
.067
-.14
3.64
1
.00
.885
1.000
-1.89
1.89
2
-2.25(*)
.885
.023
-4.14
-.36
-3.00(*)
-4.00(*)
-1.75
* The mean difference is significant at the .05 level.
.885
.885
.885
.004
.000
.067
-4.89
-5.89
-3.64
-1.11
-2.11
.14
6
3
4
5
41 | P a g e
Topic # 43: Duncan’s Multiple Range Test
Use Duncan’s multiple range test for the given data to compare all pairs of treatment means.
Assume that α = 0.05
Treatments
1
2
Blocks3
4
1
2
3
4
5
6
1
1
3
2
3
4
6
3
6
4
7
2
4
8
8
3
3
5
4
2
2
1
3
1
(Int to Stat-theory, P-II) Ex # 20.8, Pg # 328
From the menus choose:
Analyze
General Linear Model
Univariate …
Values
Duncan
N
Treatment
1
Subset
1
2
3
1
4
1.75
6
4
1.75
5
4
3.50
2
4
4.00
4.00
3
4
4.75
4.75
4
4
Sig.
3.50
5.75
.079
.200
.079
Means for groups in homogeneous subsets are displayed.
The error term is Mean Square (Error) = 1.567.
Alpha = .05.
Topic # 44: Tukey’s Test
To illustrate Tukey’s test, we use the data from the cotton weight percentage experiment in the
given data. With α = 0.05 and f = 20 degree of freedom for error.
Observed Tensile Length
Weight
of Cotton
1
2
3
4
5
15
7
7
15
11
9
20
12
17
12
18
18
25
14
18
18
19
19
30
19
25
22
19
23
35
7
10
11
15
11
(Design & experiment) Ex # 3.7, Pg # 97
42 | P a g e
From the menus choose:
Analyze
General Linear Model
Univariate …
Multiple Comparisons
(I) weights
15
20
25
30
35
Mean
Difference
(I-J)
-5.60(*)
-7.80(*)
Std. Error
1.522
1.522
Sig.
.015
.001
30
-11.80(*)
1.522
.000
-16.46
-7.14
35
-1.00
1.522
.963
-5.66
3.66
15
5.60(*)
1.522
.015
.94
10.26
25
30
35
-2.20
-6.20(*)
4.60
1.522
1.522
1.522
.609
.007
.054
-6.86
-10.86
-.06
2.46
-1.54
9.26
15
7.80(*)
1.522
.001
3.14
12.46
20
2.20
1.522
.609
-2.46
6.86
30
35
-4.00
6.80(*)
1.522
1.522
.112
.003
-8.66
2.14
.66
11.46
15
20
11.80(*)
6.20(*)
1.522
1.522
.000
.007
7.14
1.54
16.46
10.86
25
4.00
1.522
.112
-.66
8.66
35
10.80(*)
1.522
.000
6.14
15.46
15
1.00
1.522
.963
-3.66
5.66
20
-4.60
-6.80(*)
-10.80(*)
1.522
1.522
1.522
.054
.003
.000
-9.26
-11.46
-15.46
.06
-2.14
-6.14
(J) weights
20
25
25
30
95% Confidence Interval
Upper bound
Lower bound
-10.26
-.94
-12.46
-3.14
Topic # 45: Regression analysis of 23 Factorial Design
A chemical engineer is investigating the yields of processes. Three process variables are of
interest, temperature, and pressure and catalyst concentration. Each variable can be run at low
and high level, and the engineer decides to run a 23 design with four centers points.
43 | P a g e
Run
1
2
3
4
5
6
7
8
9
10
11
12
Temperature
120
160
120
160
120
160
120
160
140
140
140
140
Pressure
40
40
80
80
40
40
80
80
60
60
60
60
concentration
15.0
15.0
15.0
15.0
30.0
30.0
30.0
30.0
22.5
22.5
22.5
22.5
Yields
32
46
57
65
36
48
57
68
50
44
53
56
(Design & experiment) Ex # 10.2, Pg # 402
Note: First of all we will make the coded variables of the given variables temperature as 𝑋1 ,
pressure as 𝑋2 , concentration as 𝑋3 . By using the formula 𝑋1 =
𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒−60
20
and 𝑋3 =
𝑐𝑜𝑛𝑐−22.5
From the menus choose:
Transform
Compute Variable …
7.5
.
𝑡𝑒𝑚𝑝−140
20
, 𝑋2 =
Analyze
Reports
Case Summaries …
2^4 FACTORIAL DATA
x1
x2
x3
yield
1
-1
-1
-1
32
2
1
-1
-1
46
3
-1
1
-1
57
4
1
1
-1
65
5
-1
-1
1
36
6
1
-1
1
48
7
-1
1
1
57
8
1
1
1
68
9
0
0
0
50
10
0
0
0
44
11
0
0
0
53
12
0
0
0
56
-1
-1
-1
32
Total
44 | P a g e
From the menus choose:
Analyze
Regression
Linear …
Coefficients
Unstandardized
Coefficients
Model
1
(Constant)
x1
x2
x3
B
51.000
5.625
10.625
1.125
Std. Error
.966
1.183
1.183
1.183
Standardized
Coefficients
Beta
.449
.848
.090
t
B
52.783
4.753
8.979
.951
Sig.
Std. Error
.000
.001
.000
.370
Topic # 46: Dennett’s Test (3T)
To illustrate Dennett’s test, consider the experiment development engineer is interested in
determining if the cotton weight % in a synthetic fiber affect the tensile strength and she has run
a completely randomized experiment with 5 levels of cotton weight % and 5 replicates.
Weight %
of Cotton
15
20
25
30
35
1
7
12
14
19
7
Observed Tensile Strength
2
3
4
7
15
11
17
12
18
18
18
19
25
22
19
10
11
15
5
9
18
19
23
11
Perform the analysis of Dennett’s test at 5% level of significance.
(Design & experiment) Ex # 3.10, Pg # 103
From the menus choose:
Analyze
General Linear Model
Univariate … (post hoc test)
45 | P a g e
Multiple Comparisons
Dunnett’s T3
(I) weights
1
2
3
4
5
Mean
Difference
(I-J)
-5.60
-7.80(*)
Std. Error
2.049
1.761
Sig.
.180
.026
Lower bound
-13.11
-14.60
Upper bound
1.91
-1.00
4
-11.80(*)
1.897
.003
-18.86
-4.74
5
-1.00
1.970
1.000
-8.26
6.26
1
5.60
2.049
.180
-1.91
13.11
3
4
5
-2.20
-6.20
4.60
1.679
1.822
1.897
.845
.074
.273
-8.61
-12.93
-2.36
4.21
.53
11.56
1
7.80(*)
1.761
.026
1.00
14.60
2
2.20
1.679
.845
-4.21
8.61
4
-4.00
1.490
.197
-9.53
1.53
5
6.80(*)
11.80(*)
6.20
1.581
1.897
1.822
.025
.003
.074
.86
4.74
-.53
12.74
18.86
12.93
3
4.00
1.490
.197
-1.53
9.53
5
10.80(*)
1.732
.002
4.45
17.15
1
1.00
1.970
1.000
-6.26
8.26
2
-4.60
-6.80(*)
-10.80(*)
1.897
1.581
1.732
.273
.025
.002
-11.56
-12.74
-17.15
2.36
-.86
-4.45
(J) weights
2
3
1
2
3
4
95% Confidence Interval
Topic # 47: Test of Equality of Variance (Bartlett’s)
We can apply the Bartlett’s test to the tensile strength data from the cotton weight percentage
experiment in the following data.
Weight %
of Cotton
15
20
25
30
35
1
7
12
14
19
7
Observed Tensile Strength
2
3
4
7
15
11
17
12
18
18
18
19
25
22
19
10
11
15
5
9
18
19
23
11
(Design & experiment) Ex # 3.4, Pg # 82
From the menus choose:
Analyze
Compare Means
One Way ANOVA … (option)
46 | P a g e
Test of Homogeneity of Variances
Levine
Statistic
.939
df1
df2
4
Sig.
.462
20
Topic # 48: Effect Size for the One-Way (Eta-Square)
Compute the Eta-Square for the following data shows version of the musical experiment of the
five groups.
1
16
16
14
13
12
Groups
3
26
24
22
20
20
2
23
21
20
20
17
4
11
9
7
7
7
5
10
9
9
6
6
(Statistics bhvrl Scnc) Pg # 304
From the menus choose:
Analyze
General Linear Model
Univariate … (option/effect size)
Tests of Between-Subjects Effects
Source
Corrected
Model
Intercept
Groups
Type III
Sum of
Squares
Mean
Square
df
884.400(a)
4
221.100
5329.000
1
5329.000
884.400
4
221.100
Error
85.600
20
4.280
Total
6299.000
25
970.000
24
Corrected
Total
F
51.659
1245.09
3
51.659
Sig.
Partial Eta
Squared
.000
.912
.000
.984
.000
.912
Topic # 49: Control Chart by the Cases is Units
A food company puts mango juice into cans advertised as containing 10 ounces, of the juice. The
weights of the juice drained from cans immediately after filling for 20 samples are taken by
random method. Construct control chart.
47 | P a g e
Sample Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Weight of each can ( 4 cans in each sample, n = 4)
15
12
13
20
10
8
8
14
8
15
17
10
12
17
11
12
18
13
15
4
20
16
14
20
15
19
23
17
13
23
14
16
9
8
18
5
6
10
24
20
5
12
20
15
3
15
18
18
6
18
12
10
12
9
15
18
15
15
6
16
18
17
8
15
13
16
5
4
10
20
8
10
5
15
10
12
6
14
33
14
(Txt Busi Stat) Ex # 4, Pg # 32
From the menus choose:
Analyze
Quality Control
Control Charts …
48 | P a g e
Topic # 50: Control Chart by the Sub-Groups
The tables below gives the measurements obtain in 20 samples (subgroups). Constructs the
control charts based on the mean and the range. The values of these statistics are given below for
the respective samples.
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20
-1 2
1
2
1
1
-1 1
2
-2
0
2
0
0
-1
1
2
2
0
3
2
0
1
1
-1 -1 1
1
1
1
1
1
1
0
2
-1
1
0
2
-3
1
1
0
0
0
2
0
2
-1 -2
-3
-1
-3
-1
1
2
-1
1
1
-1
0
0
0
-1 0
0
-2 -1 0
2
2
0
2
0
1
0
0
0
-1
1
1
1
1
0
1
1
0
1
1
2
-2
0
1
1
2
-1 -2 1
0
0
(Txt Busi Stat) Ex # 1, Pg # 28
49 | P a g e
From the menus choose:
Analyze
Quality Control
Control Charts …
Topic # 51: Second Degree Parabola
Fit a Second degree parabola to the following data, taking X as an independent variable.
X
Y
0
1
1
1.8
2
3
4
1.3
2.5
6.3
(Int to Stat-theory, P-I) Ex # 12.2, Pg # 489
50 | P a g e
From the menus choose:
Analyze
Regression
Curve Estimation …
Model Summary and Parameter Estimates
Model Summary
Equation
Quadratic
R Square
.915
F
Parameter Estimates
df1
10.735
df2
2
Sig.
2
1
1.6
2
4.5
3
13.8
4
40.2
B1
1.420
.085
Topic # 52: Exponential Curve
Fit an Exponential curve 𝑌 = 𝑎𝑒 𝑏𝑋 to the following data.
X
Y
Constant
b2
-1.070
.550
5
6
125.0 363.0
(Int to Stat-theory, P-I) Ex # 12.5, Pg # 495
From the menus choose:
Analyze
Regression
Curve Estimation …
Model Summary and Parameter Estimates
Model Summary
Equation
Exponential
R Square
1.000
F
32439.854
df1
Parameter Estimates
df2
1
Sig.
.000
4
Constant
.524
b1
1.090
Topic # 53: Third Degree Parabola
Fit a Third degree parabola to the following data, taking X as an independent variable.
X
Y
0
1
1
1.8
2
1.3
3
2.5
4
6.3
(Int to Stat-theory, P-I) Prb # 12.12, Pg # 506
From the menus choose:
Analyze
Regression
Curve Estimation …
51 | P a g e
Model Summary and Parameter Estimates
Model Summary
Equatio
n
Cubic
R
Square
F
.997
df1
98.016
Parameter Estimates
df2
3
Constan
t
Sig.
1
1.030
.074
b1
b2
1.725
b3
-1.400
.325
Topic # 54: Power Curve
Fit the curve of the from Y = 𝑎𝑋 𝑏 to the following data on the unit cost in dollars of producing
certain electronic component’s and the number of units produced
Lot size (X)
50
100
250
500
1000
Unit cost (Y)
108
53
24
9
5
Use the result to estimate the unit cost for a lot of 400 components.
(Int to Stat-theory, P-I) Prb # 12.21, Pg # 507
From the menus choose:
Analyze
Regression
Curve Estimation …
Model Summary and Parameter Estimates
Equation
Model Summary
R Square
Power
.994
F
530.639
df1
Parameter Estimates
df2
1
Sig.
3
.000
Constant
6480.770
b1
-1.040
Case Summaries
1
2
3
4
5
6
Total
N
lot size
50
100
250
500
1000
400
6
cost
108
53
24
9
5
5
Fit for cost with
lot size from
CURVEFIT,
MOD_2 POWER
110.86627
53.91955
20.79277
10.11252
4.91820
12.75381
6
52 | P a g e
Topic # 55: Testing Hypothesis about P’s of The Multinomial Distribution
Two hundred were chosen at random from a set of table. The frequencies of the digits were.
Digits
Frequency
0
18
1
19
2
23
3
21
4
16
5
25
6
22
7
20
8
21
9
15
(Int to Stat-theory, P-II) Ex # 17.8, Pg # 192
From the menus choose:
Data
Weight cases …
Analyze
Non-Parametric Tests
Chi-Square …
Test Statistics
Digits
ChiSquare(a)
df
Asymp. Sig.
4.300
9
.891
Topic # 56: Goodness of Fit Test
Five pennies were tossed 1000 times and the number of heads observed as given below
Number of heads
Frequency
0
38
1
144
2
342
3
287
4
164
5
25
Test whether a Binomial distribution given as satisfactory fit to this data while the p = 0.494.
(Int to Stat-theory, P-II) Ex # 17.10, Pg # 196
From the menus choose:
Transform
Compute Variable
Make a target variable “Expected” and fit Binomial distribution.
Data
Weight cases …
Analyze
Non-Parametric Tests
Chi-Square …
Values (Add all values from the expected column)
53 | P a g e
Num_of_Heads
0
Observed N
38
Expected N
33.2
1
144
161.9
-17.9
2
342
316.2
25.8
3
287
308.7
-21.7
4
164
150.7
13.3
25
29.4
-4.4
5
Total
Residual
4.8
1000
Test Statistics
Num_of_Heads
ChiSquare(a)
df
8.146
5
Asymp. Sig.
.148
Topic # 57: Testing Hypothesis about Equality of Several Proportion
Form the adult male population of seven large cities random samples of sizes indicated below
were taken, and the numbers of married and single man recorded.
City
Married
Single
Total
A
133
36
169
B
164
57
221
C
155
40
195
D
106
37
143
E
153
55
208
F
123
33
156
G
146
36
182
Total
980
294
1274
(Int to Stat-theory, P-II) Ex # 17.21, Pg # 218
From the menus choose:
Data
Weight cases …
Analyze
Descriptive Statistics
Crosstabs …
54 | P a g e
Chi-Square Tests
Pearson Chi-Square
Likelihood Ratio
Linear-by-Linear
Association
N of Valid Cases
Value
5.337
5.327
6
6
Asymp. Sig.
(2-sided)
.501
.503
1
.617
df
.250
1274
Topic # 58: The Chi-Square Test as Test of Homogeneity
In certain community, a random sample of 50 men and another sample of 50 women over 21
year of age were asked about their eruptional background, classified as junior high, senior high
or college. The results are,
Junior High
13
23
Male
Female
Senior High
25
20
College
12
7
(Int to Stat-theory, P-II) Ex # 17.22, Pg # 220
From the menus choose:
Data
Weight cases …
Analyze
Descriptive Statistics
Crosstabs …
Chi-Square Tests
Pearson Chi-Square
Likelihood Ratio
Linear-by-Linear
Association
N of Valid Cases
Value
4.649
4.703
Asymp. Sig.
(2-sided)
df
4.275
2
2
.098
.095
1
.039
100
Topic # 59: Fit the Second Degree Curve Trend
Fit the second degree trend curve (parabola) to the following data and compute the trend values.
Year
Index (whole sale price)
Coded
1931
96
-7
1933
87
-5
1935
91
-3
1937
102
-1
1939
108
1
1941
139
3
1943
307
5
1945
289
7
(Int to Stat-theory, P-I) Ex # 13.7, Pg # 527
55 | P a g e
From the menus choose:
Analyze
Regression
Curve Estimation …
Model Summary and Parameter Estimates
Model Summary
Equation
Quadratic
R Square
.874
F
df1
17.314
Parameter Estimates
df2
2
Sig.
5
.006
Constant
110.219
b1
15.482
b2
2.007
Topic # 60: Growth Model
In the exercise 3.22 we presented data on U.S. real GDP for the period 1972-1991 suppose we
want to find out the rate of growth of real GDP in this period. Let 𝑌𝑡 = real GDP (RGDP) at time t
and 𝑌0 = the initial (1972) value of real GDP.
Time
1972
1973
1974
1975
1976
1977
1978
GDP(1987) 3,107.1 3,268.6 3,248.1 3,221.7 3,380.8 3,533.3 3,703.5
Time
1979
1980
1981
1982
1983
1984
1985
GDP(1987) 3,796.8 3,776.3 3,843.1 3,760.3 3,906.6 4,148.5 4,279.8
Time
1986
1987
1988
1989
1990
1991
GDP(1987) 4,404.5 4,539.9 4,718.6 4,838.0 4,877.5 4,821.0
(Basic Eco) Sec # 6.5, Pg # 169
From the menus choose:
Analyze
Regression
Curve Estimation …
Model Summary and Parameter Estimates
Model Summary
Equation
Growth
R Square
.974
F
668.959
df1
Parameter Estimates
df2
1
18
Sig.
.000
Constant
8.014
b1
.025
Topic # 61: Autocorrelation Function for 25 Lags (ACF)
Calculate the autocorrelation function for the GDP of United States, 1970 – I to 1991 - IV by
using the 25 lags from the following data.
56 | P a g e
Quarter
1970 – I
1970 – II
1970 – III
1970 – IV
1971 – I
1971 – II
1971 – III
1971 – IV
1972 – I
1972 – II
1972 – III
1972 – IV
1973 – I
1973 – II
1973 – III
1973 – IV
1974 – I
1974 – II
1974 – III
1974 – IV
1975 – I
1975 – II
GDP
2872.8
2860.3
2896.6
2873.7
2942.9
2947.4
2966.0
2980.8
3037.3
3089.7
3125.8
3175.5
3253.3
3267.6
3264.3
3289.1
3259.4
3267.6
3239.1
3226.4
3154.0
3190.4
Quarter
1975 – III
1975 – IV
1976 – I
1976 – I1
1976 – I1I
1976 – IV
1977 – I
1977 – II
1977 – III
1977 – IV
1978 – I
1978 – II
1978 – III
1978 – IV
1979 – I
1979 – II
1979 – III
1979 – IV
1980 – I
1980 – II
1980 – III
1980 – IV
GDP
3249.9
3292.5
3356.7
3369.2
3381.0
3416.3
3466.4
3525.0
3574.4
3567.2
3591.8
3707.0
3735.6
3779.6
3780.8
3784.3
3807.5
3814.6
3830.8
3732.6
3733.5
3808.5
Quarter
1981 – I
1981 – II
1981 – III
1981 – IV
1982 – I
1982 – II
1982 – III
1982 – IV
1983 – I
1983 – II
1983 – III
1983 – IV
1984 – I
1984 – II
1984 – III
1984 – IV
1985 – I
1985 – II
1985 – III
1985 – IV
1986 – I
1986 – II
GDP
3860.5
3844.4
3864.5
3803.1
3756.1
3771.1
3754.4
3759.6
3783.5
3886.5
3944.4
4012.1
4089.5
4144.0
4166.4
4194.2
4221.8
4254.8
4309.0
4333.5
4390.5
4387.7
Quarter
1986 – III
1986 – IV
1987 – I
1987 – II
1987 – III
1987 – IV
1988 – I
1988 – II
1988 – III
1988 – IV
1989 – I
1989 – II
1989 – III
1989 – IV
1990 – I
1990 – II
1990 – III
1990 – IV
1991 – I
1991 – II
1991 – III
1991 – IV
GDP
4412.6
4427.1
4460.0
4515.3
4559.3
4625.5
4655.3
4704.8
4734.5
4779.7
4809.8
4832.4
4845.6
4859.7
4880.8
4900.3
4903.3
4855.1
4824.0
4840.7
4862.7
4868.0
(Basic Eco) Sec # 21.3, Pg # 716
From the menus choose:
Analyze
Time Series
Autocorrelations …
57 | P a g e
Autocorrelations
Autocorrel
ation
Lag
1
Value
.969
Std.
Error(a)
df
Box-Ljung Statistic
Sig.(b)
Value
df
.105
85.462
1
.000
2
.935
.104
166.016
2
.000
3
.901
.104
241.717
3
.000
4
.866
.103
312.393
4
.000
5
.830
.102
378.100
5
.000
6
.791
.102
438.566
6
.000
7
.752
.101
493.849
7
.000
8
.713
.101
544.108
8
.000
9
.675
.100
589.773
9
.000
10
.638
.099
631.121
10
.000
11
.601
.099
668.328
11
.000
12
.565
.098
701.649
12
.000
13
.532
.097
731.556
13
.000
14
.500
.097
758.290
14
.000
15
.468
.096
782.020
15
.000
16
.437
.095
803.025
16
.000
17
.405
.095
821.346
17
.000
18
.375
.094
837.237
18
.000
19
.344
.093
850.786
19
.000
20
.313
.093
862.172
20
.000
21
.279
.092
871.390
21
.000
22
.246
.091
878.647
22
.000
23
.214
.091
884.216
23
.000
24
.182
.090
888.314
24
.000
25
.153
.089
891.246
25
.000
Topic # 62: Partial Autocorrelation Function for 25 Lags (PACF)
Calculate the partial autocorrelation function for the GDP of United States, 1970 – I to 1991 - IV
by using the 25 lags from the following data.
58 | P a g e
Quarter
1970 – I
1970 – II
1970 – III
1970 – IV
1971 – I
1971 – II
1971 – III
1971 – IV
1972 – I
1972 – II
1972 – III
1972 – IV
1973 – I
1973 – II
1973 – III
1973 – IV
1974 – I
1974 – II
1974 – III
1974 – IV
1975 – I
1975 – II
GDP
2872.8
2860.3
2896.6
2873.7
2942.9
2947.4
2966.0
2980.8
3037.3
3089.7
3125.8
3175.5
3253.3
3267.6
3264.3
3289.1
3259.4
3267.6
3239.1
3226.4
3154.0
3190.4
Quarter
1975 – III
1975 – IV
1976 – I
1976 – I1
1976 – I1I
1976 – IV
1977 – I
1977 – II
1977 – III
1977 – IV
1978 – I
1978 – II
1978 – III
1978 – IV
1979 – I
1979 – II
1979 – III
1979 – IV
1980 – I
1980 – II
1980 – III
1980 – IV
GDP
3249.9
3292.5
3356.7
3369.2
3381.0
3416.3
3466.4
3525.0
3574.4
3567.2
3591.8
3707.0
3735.6
3779.6
3780.8
3784.3
3807.5
3814.6
3830.8
3732.6
3733.5
3808.5
Quarter
1981 – I
1981 – II
1981 – III
1981 – IV
1982 – I
1982 – II
1982 – III
1982 – IV
1983 – I
1983 – II
1983 – III
1983 – IV
1984 – I
1984 – II
1984 – III
1984 – IV
1985 – I
1985 – II
1985 – III
1985 – IV
1986 – I
1986 – II
GDP
3860.5
3844.4
3864.5
3803.1
3756.1
3771.1
3754.4
3759.6
3783.5
3886.5
3944.4
4012.1
4089.5
4144.0
4166.4
4194.2
4221.8
4254.8
4309.0
4333.5
4390.5
4387.7
Quarter
1986 – III
1986 – IV
1987 – I
1987 – II
1987 – III
1987 – IV
1988 – I
1988 – II
1988 – III
1988 – IV
1989 – I
1989 – II
1989 – III
1989 – IV
1990 – I
1990 – II
1990 – III
1990 – IV
1991 – I
1991 – II
1991 – III
1991 – IV
GDP
4412.6
4427.1
4460.0
4515.3
4559.3
4625.5
4655.3
4704.8
4734.5
4779.7
4809.8
4832.4
4845.6
4859.7
4880.8
4900.3
4903.3
4855.1
4824.0
4840.7
4862.7
4868.0
(Basic Eco) Sec # 21.3, Pg # 716
From the menus choose:
Analyze
Time Series
Autocorrelations …
59 | P a g e
Partial Autocorrelations
Partial
Autocorrel
ation
.969
Std. Error
.107
2
-.058
.107
3
-.020
.107
4
-.045
.107
5
-.024
.107
6
-.062
.107
7
-.029
.107
8
-.024
.107
9
.009
.107
10
-.010
.107
11
-.020
.107
12
-.012
.107
13
.020
.107
14
-.012
.107
15
-.021
.107
16
-.001
.107
17
-.041
.107
18
-.005
.107
19
-.038
.107
20
-.017
.107
21
-.066
.107
22
-.019
.107
23
-.008
.107
24
-.018
.107
25
.017
.107
Lag
1
Topic # 63: Time Series Graph
Plot the time series graph for the four quarters of each year of 1983 to 1986. Data is given the
following table.
Year
1983
1984
1985
1986
I
20
15
15
18
Quarter
II
III
26
43
21
43
32
45
36
50
IV
57
68
73
78
(Stat for Eco & Mngmnt) Ex # 14.1, Pg # 352
From the menus choose:
Analyze
Time Series
Sequence Charts …
60 | P a g e
80
values
60
40
20
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
quarter
Topic # 64: Exponential Smoothing
The data from 1985 January to 1987 December the total 36 observations are displayed in the
following table;
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1985 1.3 .8 1.2
.9
.9
1.1 .8
.9
.9 1.1 .9
1.3
1986 1.0 1.0 1.1 1.6 1.1 1.0 1.2 1.0
.9 1.0 1.0 1.1
1987 .8 1.0 1.0 1.2 1.1 1.3 1.1 1.1 1.3 1.2 1.3 1.1
(Busi Stat) Sec # 7.6.1, pg # 154
From the menus choose:
Transform
Create Time Series … (smoothing)
Analyze
Time Series
Sequence Charts …
61 | P a g e
THE END
62 | P a g e