Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
• When 2
S
FSTAT 1
2
df1 = n1 – 1 ; df2 = n2 – 1
S 2
• In the F table,
– numerator degrees of freedom determine the column
– denominator degrees of freedom determine the row
03-Oct-18 F distribution and ANNOVA 3
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Finding the Rejection Region
H0: σ12 = σ22 H0: σ12 ≤ σ22
H1: σ12 ≠ σ22 H1: σ12 > σ22
/2
/2
0 F 0 F
Reject H0 Do not Reject H0 Do not Reject H0
Fα/2
reject H0
F(1- α/2)
reject H0
Fα
Critical values: F(1-α/2, n2-1, n1-1)=?
F(α/2, n1-1,n2-1) =?
Reject H0 if FSTAT > Fα
Rejection region:
Reject H0 if FSTAT < F(1-α/2,n2-1,n1-1) or
FSTAT > F(α/2,n1-1,n2-1)
03-Oct-18 F distribution and ANNOVA 4
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
F Test: An Example
You are a financial analyst for a brokerage firm. You want to
compare dividend yields between stocks listed on the NYSE &
NASDAQ. You collect the following data:
NYSE NASDAQ
Number 21 25
Mean 3.27 2.53
Std dev 1.30 1.16
fdist 17
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Example:continued
H0: s12 = s22
H a: s12 > s22
fdist 18
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
One-Way Analysis Of Variance
(ANOVA) Setting
• Want to examine differences among more than two groups
• The groups involved are classified according to levels of a
factor of interest (numerical or categorical)
– Different levels produce different groups
– Think of each group as a sample from a different
population
• Observe effects on the dependent variable
– Are the groups the same?
• When there is only 1 factor the design is called a completely
randomized design
03-Oct-18 F distribution and ANNOVA 19
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
One-Way Analysis of Variance
• Evaluate the difference among the means of three or
more groups
Examples: Accident rates for 1st, 2nd, and 3rd shift
Expected mileage for five brands of tires
• Assumptions
– Populations are normally distributed
– Populations have equal variances
– Samples are randomly and independently drawn
• H0 : μ1 μ2 μ3 μc
– All population means are equal
– i.e., no factor effect (no variation in means
among groups)
H1 : Not all of the population means are the same
•
– At least one population mean is different
– i.e., there is a factor effect
– Does not mean that all population means are
03-Oct-18 different (some pairs
F distribution may be the same)
and ANNOVA 21
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
One-Way ANOVA
H0 : μ1 μ2 μ3 μc
H1 : Not all μj are the same
The Null Hypothesis is True
All Means are the same:
(No Factor Effect)
03-Oct-18
μ 2andANNOVA
μF1distribution μ3 22
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
One-Way ANOVA
(continued)
H0 : μ1 μ2 μ3 μc
H1 : Not all μj are the same
The Null Hypothesis is NOT true
At least one of the means is different
(Factor Effect is present)
or
μ1 μ2 μ3 μ1 μ2 μ3
03-Oct-18 F distribution and ANNOVA 23
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Partitioning the Variation
SST ( X ij X ) 2
Where: j 1 i 1
Response, X
j 1
Where:
SSA = Sum of squares among groups
c = number of groups
nj = sample size from group j
Xj = sample mean from group j
X = grand mean (mean of all data values)
03-Oct-18 F distribution and ANNOVA 29
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Among-Group Variation
(continued)
c
SSA n j ( X j X ) 2
j 1
SSA
Variation Due to
Differences Among Groups MSA
c 1
Mean Square Among =
SSA/degrees of freedom
i j
03-Oct-18 F distribution and ANNOVA 30
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Among-Group Variation
(continued)
SSA n1 ( X 1 X )2 n2 ( X 2 X )2 nc ( X c X )2
Response, X
X3
X2 X
X1
SSW ( X ij X j ) 2
j 1 i 1
Where:
SSW = Sum of squares within groups
c = number of groups
nj = sample size from group j
Xj = sample mean from group j
03-Oct-18
Xij = ith observation in group j
F distribution and ANNOVA 32
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Within-Group Variation
(continued)
c nj
SSW ( X ij X j )2
j 1 i 1
μj
03-Oct-18 F distribution and ANNOVA 33
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Within-Group Variation
(continued)
Response, X
X3
X2
X1
c = number of groups
n = sum of the sample sizes from all groups
df = degrees of freedom
03-Oct-18 F distribution and ANNOVA 36
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
One-Way ANOVA
F Test Statistic
H0: μ1= μ2 = … = μc
H1: At least two population means are different
• Degrees of freedom
– df1 = c – 1 (c = number of groups)
– df2 = n – c (n = sum of sample sizes from all populations)
03-Oct-18 F distribution and ANNOVA 37
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Interpreting One-Way ANOVA
F Statistic
• The F statistic is the ratio of the among
estimate of variance and the within
estimate of variance
– The ratio must always be positive
– df1 = c -1 will typically be small
– df2 = n - c will typically be large
Decision Rule:
Reject H0 if FSTAT > Fα,
otherwise do not reject
H0 0 Do not Reject H0
reject H0
Critical Decision:
Value:
Reject H0 at = 0.05
Fα = 3.89
= .05 Conclusion:
There is evidence that
0 Do not Reject H
0
at least one μj differs
reject H0
03-Oct-18
Fα = 3.89
FSTAT F=distribution
25.275 from the rest
and ANNOVA 42
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
The Tukey-Kramer Procedure
μ1= μ2 μ3 x
MSW 1 1
Critical Range Q α
2 n j n j'
where:
Qα = Upper Tail Critical Value from Studentized
Range Distribution with c and n - c degrees
of freedom
MSW = Mean Square Within
nj and nj’ = Sample sizes from groups j and j’
Q α 3.77
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
03-Oct-18 F distribution and ANNOVA 46
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
The Tukey-Kramer Procedure:
Example
(continued)
3. Compute Critical Range:
MSW 1 1 93.3 1 1
Critical Range Q α 3.77 16.285
2 n j n j' 2 5 5
4. Compare:
5. All of the absolute mean x1 x 2 23.2
differences are greater than critical
range. Therefore there is a significant x1 x 3 43.4
difference between each pair of
means at 5% level of significance. x 2 x 3 20.2
Thus, with 95% confidence we can conclude
that the mean distance for club 1 is greater
than club 2 and 3, and club 2 is greater than
club 3.
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
ANOVA Assumptions
22 19 18 21
22.5 19.5 17 20 X1 22 X 2 19.5833
21.5 19 18.5 21.5
X 3 17.6667 X 4 20.5833
22 20 17 20
22.5 19 18.5 21 X 19.95833
21.5 21 17 20
22 19 18 21
22.5 19.5 17 20
21.5 19 18.5 21.5
22 20 17 20
22.5 19 18.5 21
21.5 21 17 20
SSW = (22 – 22)2 + (22.5 – 22)2 + (21.5 – 22)2 + (22 – 22)2 + (22.5 – 22)2
+ (21.5 – 22)2 + (22.5 – 19.5833)2 + (22.5 – 19.5833)2 + (22.5 – 19.5833)2
+ (22.5 – 19.5833)2 + (22.5 – 19.5833)2 + (22.5 – 19.5833)2 + (18 –
17.6667)2 + (17 – 17.6667)2 + (18.5 – 17.6667)2 + (17 – 17.6667)2 + (18.5
– 17.6667)2 + (17 – 17.6667)2 + (21 – 20.5833)2 + (20 – 20.5833)2 + (21.5
– 20.5833)2 + (20 – 20.5833)2 + (21 – 20.5833)2 + (20 – 20.5833)2 = 9.25
03-Oct-18 F distribution and ANNOVA 52
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Solution: ANOVA Table
Critical Decision:
Value:
Reject H0 at = 0.05
Fα = 3.10
= .05 Conclusion:
There is evidence that
0 Do not Reject H
0
at least one μj differs
reject H0
FSTAT = 43.03 from the rest
Fα = 3.10
03-Oct-18 54
F distribution and BITS
ANNOVA
Pilani, Deemed to be University under Section 3 of UGC Act, 1956
ANOVA: EXCEL