Professional Documents
Culture Documents
Sidm Tee Formula
Sidm Tee Formula
From q From q
posterior Probability
P(Ai|B)
#DIV/0!
#DIV/0!
No of success x 4
No of trials n 6
Probability of successp 0.2
Cummulative for x success false, for x<=success true
probability 0.01536
probability 0.180447
-0.999943 5.684024E-05
Formula norm.dist
formula inverse 47.25503 norm.inv What score would be needed to be in the 85th percentile?
e 85th percentile?
Type I Error
Type II Error
There
is a tradeoff between these two errors The probability of making one type
of error can be reduced only if we are willing to increase probability of making the
other type of error
bility of making the
Z value t value
n 36
μ 1.5
x bar 1.4
σ 0.12
Z=(x bar-μ)/SE -5
dof 19
p value (Two tailed)
At 0.95 1.7291328115
n 60
P bar 0.35
Po 0.46 1-Po 0.54
P bar-Po -0.11
Po*(1-Po)/n 0.00414
Sq root(Po*(1-Po)/n) 0.064343 standard error
z -1.709592
2.326348
n- sample size 70
p bar 0.65
q bar 0.35
standard error
root (p bar * q bar/ n)
z cv/ t cv formula for t t.inv.2t
confidence level
margin og error
lower limit
p bar- margin of error
Upper limit
p bar + margin of error
1. A study was undertaken to determine customer satisfaction in Canadian automobile markets following certain
z cv 1.644854
aplha 0.05
P value 0.895065
0.104935
if p< alpha we reject h0
ile markets following certain changes in customer service. Suppose that before the changes, the average customer satisfaction rating, on a
stomer satisfaction rating, on a scale of 0 to 100, was 77. A survey questionnaire was sent to a random sample of 350 residents who bough
le of 350 residents who bought new cars after the changes in customer service were instituted, and the average satisfaction rating for this
rage satisfaction rating for this sample was found to be x = 84; the sample standard deviation was found to be s=28. Use an α of your choic
e s=28. Use an α of your choice, and determine whether statistical evidence indicates a change in customer satisfaction. If you determine
satisfaction. If you determine that a change did occur, state whether you believe customer satisfaction has improved or deteriorated.
mproved or deteriorated.
Confidence Interval calculation (z)
Sample 1 Sample 2
n1 50 n2 35
x1 bar 13.6 x2 bar 11.6
sigma 1 2.2 sigma 2 3
CI
90%
Point estimator x1 bar-x2 bar 2 95%
99%
sigma1^2/n1 0.0968 sigma2^2/n2 0.257143
Minimum Maximum
Sample 1 Sample 2
n1 50 n2 35
x1 bar 13.6 x2 bar 11.6
s1 2.2 s2 3
Do 0
Point estimator x1 bar-x2 bar 3.5
Sample 1
alpha/2 z(alpha/2) n1 35
0.05 1.645 x1 bar 13.6
0.025 1.96 sigma 1 5.2
0.005 2.576
sample sd/root n
sigma1^2/n1 0.772571429
Z alpha or Z cv * standard error
z 2.179502689
Find p value using norm.s.dist
n place of sigma -formula remains the same)
s1
n1
Sample 2 s1^2/n1
n2 40
x2 bar 10.1
sigma 2 8.5
(s1^2/n1)^2
1/(n1-1)
sigma2^2/ 1.80625
Calculation for degrees of freedom
2.5 s2 4.8
20 n2 30
Numerator 1.16748025
Denominator 0.0254785612523
df 45.822063437591 45
Excel formula
Observed Frequencies (fij)
Response 1 2 3 Total
Yes 150 150 96 396
No 100 150 104 354 chisq.inv.rt
Total 250 300 200 750
p value 0.018379
Reject Ho if this value is less than 0.05
chi sq test for independence Ho: 2 attributes are independent of each other
H1: 2 attribues are not independent of each other
Chi-Square 17.03
Dof (row-1)*(column-1) 4
0.001909
Less than 0.05 so reject Ho
Type I
Ho
Ha
Population 1
Population 2
n1 26 n2 25
s1 170 s2 100
s1^2 28900 s2^2 10000
Remember that the population having a higher variance needs to be taken as numerator
F 2.89
dof(num) 25
dof(den) 24
Now, look for 2.89 in the F table with numerator dof as 25 and denominator dof as 24 and note down the prob va
p value
If less than 0.05 then reject null hypothesis
Type II
F value calculation
compare f stats and f cv
F value 2.347878
F test stat < F CV
Level of significance 0.05 Do not reject Ho
dof_numerator 10
dof_denominator 20
T test 2 sample assuming unequal varinace T test paired 2 sample for mean
ed 2 sample for mean
Develop the analysis of variance computa
A B C design. At α =_x0002_ .05, is there a signi
136 107 92
120 114 82
113 125 85 H0: muA=muB=muC
107 104 101 Ha: not same
131 107 89
114 109 117
129 97 110
102 114 120
104 98
89 106
sample (n j) 8 10 10
x(bar) 119.00 107.00 100.00 Average 107.93
variance 146.86 96.44 173.78 n (T) 28
St dev 12.12 9.82 13.18 k 3
k-1 2
Anova table
Source of variation Sum of Square dof mean sq F p value
Treatments 1617.86 2.00 808.93 5.84 0.01
Errors 3460.00 25.00 138.40
Total 5077.86 27.00
p value 0.01
p value is less than 0.05 and hence reject HO
SUMMARY
Groups Count Sum Average Variance
A 8 952 119 146.8571
B 10 1070 107 96.44444
C 10 1000 100 173.7778
Data analysis table - Excel
ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 1617.86 2.00 808.93 5.84 0.01 3.39
Within Groups 3460.00 25.00 138.40
Case where sample size is different. Another case where sample size is same, refer q1 session 18
Equation : y=0.2+2.6*x
For SST For SSE
SST=SSR+SSE
SSR=SST-SSE 67.6
(Yi-Y hat)^2
0.04
2.56
9
0.16
0.64
dregree of frredom
SUMMARY
Groups Count Sum Average Variance
A 5 15 3 3.5
B 5 16 3.2 4.7
C 5 25 5 2.5
ANOVA
Source of VariationSS df MS F P-value F crit
Between Groups 12.13333 2 6.066667 1.700935 0.22369 3.885294
Within Groups 42.8 12 3.566667
Total 54.93333 14
F test < F critical
P value > 0.05 (alpha) : Do not reject Ho : So all schools are equally effective.
lways use WITHOUT REPLICATION!
Anova: Two-Factor Without Replication
12 Column1
23
33 Mean 21.909090909
16 Standard Error 2.2901153026
19 Median 23
20 Mode #N/A
24 Standard Deviation 7.5954531854
28 Sample Variance 57.690909091
9 Kurtosis -0.7319509305
26 Skewness -0.262651537
31 Range 24
Minimum 9
Maximum 33
Sum 241
Count 11
Confidence Level(95.0%) 5.1026948812
Column1
17.5
Mean 21.90909 27
Standard E 2.290115
Median 23
This chart isn't available in your version of Excel.
Mode #N/A
Standard D 7.595453 Editing this shape or saving this workbook into a different file form
Sample Var 57.69091 permanently break the chart.
Kurtosis -0.731951
Skewness -0.262652
Range 24
Minimum 9
Maximum 33
Sum 241
Count 11
Largest(1) 33
Smallest(1) 9
Confidence 5.102695
sion of Excel.