Formula Sheet

Ungrouped Data
Sample variance  ( x) 2 
 x   
2

 n 
s2 
n 1
Grouped Data
Sample mean k
mj  f j
j 1
x  k
 fj
j 1

Sample mode c   f mo  f mo 1 
M o  Omo 
 f mo  f mo 1    f mo  f mo 1 
Sample median 2  Fme 1 ]
c[ n
M e  O me 
f me
Sample variance   ( f j  m j ) 2  
 1  
s 2
  f j  mj 
2

 

 n 1   
n


ith Percentile  pn 
c  FPi 1 
 100 
Pi  O Pi 
f Pi

Graphical description
Skewness Me  Mo 3  x  M e 
s OR s
Probability counting methods
Permutation n!
n
Pr 
n  r !
Combination n n!
n
C r    
 r  r!n  r !
Probability
Probability Addition Law P(A or B)  P ( A  B )  P ( A)  P ( B )  P ( A  B )
Conditional Probability P( A  B)
P( A | B)  if P (B)>0
P( B)
P( A  B)
P( B | A)  if P (A)>0
P( A)
Probability Distributions
If X is binomially distributed then P( X  xi )  n C xi  p xi  (1  p) n xi
If X is normally distributed then  x 
P( X  xi )  P Z  i 
  
Confidence Intervals
Large sample (n ≥ 30)   s2 s2 
 s  ( x1  x 2 )  z   1  2 
 x  z    

  n   n1 n 2 
2  2 
Small sample (n < 30)   s   1 1
 x  t    
 ( x1  x2 )  t   s p   
 2  n   n1 n2 
(n1  1)  s12  (n2  1)  s22 2
s 2p 
n1  n2  2 γ=n-1 γ=n1+ n2 – 2

Sample proportion  pˆ (1  pˆ )   pˆ 1 (1  pˆ 1 ) pˆ 2 (1  pˆ 2 ) 
 pˆ  z     pˆ 1  pˆ 2  z    
 2
n   2
n1 n2 

Hypothesis Testing
Large sample (n ≥ 30) X  ( X 1  X 2 )  o
Z Z
s s12 s22

n n1 n2
Small sample (n < 30) X  ( X 1  X 2 )  o
T T
(n1  1)  s  (n2  1)  s
2 2
s
s 2p  1 2
sp 
1 1

n1  n2  2 n n1 n2
Sample proportion pˆ  po X1  X 2
Z pˆ 1  pˆ 2 pˆ 
po (1  po ) Z n1  n 2
1 1
n pˆ  (1  pˆ )    
 n1 n2 
Simple Linear Regression
Estimated slope SS xy
b
SS x
Estimated intercept a  y  bx
Correlation coefficient SS xy
r
SS x  SS y
Analysis of Variance
Source of variation Sum of squares Degrees of Mean square F-ratio
freedom
1  k  1
SST   n j   y j  y 
Treatments (T) k SST MST
2
MST  F
j 1
k 1 MSE
Sampling error (E) 2  n  k
SSE    yij  y j 
nj
k SSE
2
MSE 
j 1 i 1
nk
Total (T)  k nj 
2
  n 1
  yij 
 j 1 i 1 
TSS    yij  y 
nj k nj
  yij2   
k
2

j 1 i 1 j 1 i 1 n

Chi-square Test
Test Statistic k Oi  Ei 2
2  
i 1 Ei
Degrees of freedom for contingency table   r  1  c  1