FORMULA SHEET

Common Probability Distributions

Density Function f  g or F  g Mean   E  X  Variance  2  E  X   

2
Parametric Family
Discrete Distributions
n x
fX  x     p 1  p  I0,1,K ,n  x  np 1  p 
nx
Binomial np
x

 K  M  K 
  
x nx  K M K M n
f X  x     I0,1,K ,n  x 
K
Hypergeometric n n
M  M M M M 1
 
n

e  x
Poisson fX  x   I x  
x ! 0,1,K 
Continuous Distributions
1
fX  x   I x
b  a  a,b  ab b  a
2
Uniform
x a
FX  x   I  x   I  b ,   x  2 12
b  1  a,b 

1  1  x   2 
Normal fX  x   exp      2
 2  2    

100 1    % Confidence Interval Estimators for the Population Mean of a Normal Distribution

Cases Confidence Interval Estimators

   
Case 1:  2 is known  X  z 2 , X  z 2 
 n n

Case 2:  2 is unknown  S S 
 X  t 2 v  n  1 , X  t 2 v  n  1 
(and the sample size n  30 )  n n

Case 3:  2 is unknown  S S 
 X  z 2 , X  z 2 
(and the sample size n  30 )  n n

100 1    % Confidence Interval Estimators for the Population Proportion


ˆ
P  z

Pˆ 1  Pˆ
,

ˆ z
P

Pˆ 1  Pˆ  
  2
n
 2
n 
 
 
100 1    % Confidence Interval Estimators for the Population Variance

  n  1 S 2  n  1 S 2 
 2 , 2 
  v  n  1 
1 2 v  n  1 

 2

1 of 7
100 1    % Confidence Interval Estimators for  X  Y

Cases Confidence Interval Estimators

Case 1:  2
and  are
2
  X2  Y2  X2  Y2 
X Y

 
 X  Y  z 2 
n1 n2

, X  Y  z 2  
n1 n2 

known 

 1 1 1 
Case 2:  X2 and  Y2 are 
   n1 n2 
 
 X  Y  t 2 v  n1  n2  2  SP2   , X  Y  t 2 v  n1  n2  2  SP2  
1
 n1 n2

 

unknown but  X2   Y2   2
 n1  1 SX2   n2  1 SY2
where SP2 
n1  n2  2

 SX2 SY2 
 
 X  Y  t 2 v 
SX2 SY2
n

n
 
, X  Y  t 2 v 
n

n


 1 2 1 2 

Case 3:  X2 and  Y2 are  

 2 
  S X  SY  
2 2

unknown but  X2  Y2   n1 n2  

where v    2 2 
  S X2   SY2  

  n1    n2   
  
 n1  1 n2  1 

Case 4:  X2 and  Y2 are

 SX2 SY2 
unknown but n1  30 and
 
 X  Y  z 2
SX2 SY2
n

n

, X  Y  z 2
n
 
n


 1 2 1 2 
n2  30

100 1    % Confidence Interval Estimators for D   X  Y , Based on Paired Samples

 SD S 
 D  t 2 v  n  1 , D  t 2 v  n  1 D 
 n n

100 1    % Confidence Interval Estimators for p1  p2



Pˆ1 1  Pˆ1  
Pˆ2 1  Pˆ2  
Pˆ1 1  Pˆ1 Pˆ2 1  Pˆ2    

 ˆ ˆ

 P1  P2  z 2 n1

n2
ˆ ˆ 
, P1  P2  z 2  n1

n2 
 
 
100 1    % Confidence Interval Estimators for the Ratio of Two Standard Deviations,  X2  Y2

 SX2 SX2 
 , 
 SY F 2 v1  n1  1,v 2  n2  1 SY F1 2 v1  n1  1,v 2  n2  1 
2 2
 

2 of 7
Hypothesis Tests for the Population Mean
Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic Region of Rejection
Case 1:  is known
2
  0 z  z
X  0
  0 Z z  z
  0 
  0 n z  z 2

Case 2:  2 is unknown   0 t  t v  n  1

X  0
(and the sample size n  30 )   0 T 
S t  t v  n  1
  0   0 n t  t 2 v  n  1

Case 3:  2 is unknown   0 z  z

X  0
(and the sample size n  30 )   0 Z z  z
S
  0   0 n z  z 2

Hypothesis Test for the Population Proportion

Null Hypothesis Alternative Hypothesis Region of
Test Statistic
Ho Ha Rejection
Y  np0
p  p0 Z z  z
np0 1  p0 
p  p0 p  p0 z  z
p  p0 where Y is the number of successes in a random z  z 2
sample of size n

Hypothesis Test for the Population Variance

Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic Region of Rejection
  0  2  12 v  n  1
  0  2  2 v  n  1
  0  n  1 S 2
 2   02 2   2  12 2 v  n  1
 02
or
  2 2 v  n  1
2

3 of 7
Tests of Hypotheses for the Difference of Means Based on Two Independent Samples
Alternative
Null Hypothesis Ho Test Statistic Region of Rejection
Hypothesis Ha
Case 1:  X2 and  Y2 are known  X  y  d0
Z
 X Y   d 0
z  z
 X   y  d0  X  y  d0  X2  Y2 z  z
 z  z 2
 X   y  d0 n1 n2

Case 2:  X2 and  Y2 are

T 
 X Y   d 0

unknown but  2
 
2 2
1 1
t  t v  n1  n2  2 
X Y
 X  y  d0 SP2   
 X   y  d0  n1 n2 
 X  y  d0 t  t v  n1  n2  2 
 X   y  d0 where t  t 2 v  n1  n2  2 
 n1  1 S 2
  n2  1 S
2

SP2  X Y

n1  n2  2

Case 3:  X2 and  Y2 are  X  y  d0 t  t v 

unknown but  X2  Y2  X  y  d0 t  t  v 

 X   y  d0 t  t 2 v 
 X   y  d0

T 
 X Y   d 0
where

S X2 SY2  
  2 
  S X  SY  
2 2
n1 n2
  n1 
n2  
v  2 2 
  S X2   SY2  
  n1   n  

2
 
 1n  1 n 2  1 
Case 4:  X2 and  Y2 are

unknown but n1  30 and

 X  y  d0
Z
 X Y   d 0
z  z
 X  y  d0 S X2 SY2 z  z
n2  30 
 X   y  d0 n1 n2 z  z 2
 X   y  d0

Tests of Hypotheses for the Difference of Means Based on Paired Observations

Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic Region of Rejection
D  d 0 t  t v  n  1
D  d0
D  d 0 D  d 0 T  t  t v  n  1
Sd
D  d 0 n t  t 2 v  n  1

4 of 7
Tests of Hypotheses for the Difference between Two Proportions
Null Hypothesis Alternative Hypothesis Region of
Test Statistic
Ho Ha Rejection
Pˆ1  Pˆ2
Z
1 1
p1  p2  0  
Pˆ 1  Pˆ   
 1
n n2 
z  z
p1  p2  0 p1  p2  0 z  z
p1  p2  0 X Y z  z 2
where Pˆ  , X and Y are the number of
n1  n2
successes

Tests of Hypotheses for Equality of Variances

Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic Region of Rejection
 
2
X
2
Y
f  F1 v1  n1  1,v 2  n2  1
 
2
X
2
Y f  F v1  n1  1,v 2  n2  1
S2
    F  X2 f  F1 2 v1  n1  1,v 2  n2  1
2 2 2 2
X Y X Y
SY
or
f  F 2 v1  n1  1,v 2  n2  1

Chi-Square Goodness-of-Fit Test

Region of
Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic
Rejection
k
Oi2
2   n
i 1 Ei

The random variable X, The random variable does not where Oi is the observed
follows the hypothesized follow the hypothesized frequency of cases in the i th 2  2 v 
distribution distribution category
Ei is the expected number of

cases in the i th category

5 of 7
Chi-Square Test for Independence
Alternative
Null Hypothesis Ho Test Statistic Region of Rejection
Hypothesis Ha
r c Oij2
The random variables,  2   n
X and Y are not Eij
2  2 v   r  1   c  1 
i 1 j 1
X and Y, are
independent where r is the number of rows and c is the
independent
number of columns
For a 2  2 Contingency Table
2
The random variables,  n
X and Y are not n  ad  bc  
X and Y, are  2 2  2 v  1
2 
independent
independent
 a  b c  d a  c  b  d 

Pearson Product Moment Correlation Coefficient

n
 n  n 
n  X iYi    X i  Yi 
r  i 1  i 1  i 1 
 n 2  n   n 2  n  
2 2

 n  X i    X i   n Yi   Yi  
 i 1  i 1    i 1  
  i 1
Tests of Hypotheses for 

Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic Region of Rejection

  0 t  t v  n  2 
  0   0
 r  0  n2
t  t v  n  2 
t
  0 1 r 2 t  t 2  v  n  2 

Simple Linear Regression Model Estimated Regression Equation

Yi   0  1 X i   i Ŷ  b0  b1 X
Estimation Using the Method of Least Squares Sum of Square Error
n
 n
 n
 2
n  X iYi    X i  Yi 
 
n
SSE   Yi  Yˆi
b1  i 1  i 1  i 1  i 1
2
n
 n 
n X i    X i 
2
Mean Square Error
i 1  i 1 
SSE
b0  Y  b1 X MSE 
n2
100 1    % Confidence Interval Estimate for  1

 b  t v  n  2  S
1  2 b1 , b1  t 2 v  n  2 Sb1 
MSE
where Sb1  2
 n 
  Xi 
X i2   i 1 
n


i 1 n

6 of 7
100 1    % Confidence Interval Estimate for  0

b 0  t 2 v  n  2 Sb0 , b0  t 2 v  n  2 Sb0 
2
 n 
MSE   X i 
where Sb0   i 1 
2
n
 n 
n  X i2    X i 
i 1  i 1 
Tests for Regression Coefficients
Null Hypothesis Ho Alternative Hypothesis Ha Test Statistic Region of Rejection
b1
1  0 1  0 t t  t 2 v  n  2
Sb1

b0
0  0 0  0 t t  t 2 v  n  2
Sb0

ANOVA Table for One-Way Classification

Source of Variation Degrees of Freedom Sum of Squares Mean Square F Statistic

 
k 2
SSB   ni x i  x
SSB MSB
Between Treatment Groups k 1 MSB  F
i 1 k 1 MSW

 
k ni
SSW   xij  x i .
2 SSW
Within Treatment Groups N k MSW 
i 1 j 1 N k

 
k ni 2
Total N 1 SST   x ij  x
i 1 j 1

7 of 7