Scribd
Upload
420 views11 pages

Formula Sheet and Statistical Tables

This document provides formulas and definitions for common statistical tests and distributions including: - Hypothesis tests for means using t and z distributions - Hypothesis tests for proportions - Confidence intervals for means, proportions, and other parameters - Tests comparing two means or proportions such as z-tests, t-tests, and chi-square tests - Correlation coefficients, regression analysis, analysis of variance techniques - Nonparametric tests like the Wilcoxon and Kruskal-Wallis tests - Descriptive statistics like expected value and measures of variation

Uploaded by

donatchange
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
420 views11 pages

Formula Sheet and Statistical Tables

This document provides formulas and definitions for common statistical tests and distributions including: - Hypothesis tests for means using t and z distributions - Hypothesis tests for proportions - Confidence intervals for means, proportions, and other parameters - Tests comparing two means or proportions such as z-tests, t-tests, and chi-square tests - Correlation coefficients, regression analysis, analysis of variance techniques - Nonparametric tests like the Wilcoxon and Kruskal-Wallis tests - Descriptive statistics like expected value and measures of variation

Uploaded by

donatchange
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Formula Sheet

Hypothesis Tests for the Mean, t distribution

Mean of a uniform distribution


a+b
=
2

x
t=
s
n

Standard deviation
distribution

of

(b a )2
12

uniform Hypothesis Tests for Proportions


Ps P
z=
P (1 P )
n

Uniform probability distribution


1
P ( x) =
,
ba
if a x b and 0 elsewhere

The Standardized Normal Distribution


x
z=

Test statistics for difference between two


large sample means

x1 x 2 (1 2 )

z=
2

s1
s
+ 2
n1
n2

The confidence interval for 1 2 ,large


sample
2
2
s2
s1
+
x1 x2 z

n1 n2

z-value for Sampling Distribution of the Pooled variance


2
2
Sample Mean
( n 1) s1 + ( n 2 1) s 2
2

sp = 1
n1 + n 2 2
x
z=
s

n
Confident Interval for z-Distribution

x z
2 n

Two sample test of means-small samples



x1 x 2 ( 1 2 )

t=
1
1
s 2 p +
n1 n 2

Confident Interval for t-Distribution

s
x t

, df
n
2

The confidence interval for 1 2 , small


samples


s 2 p 1 + 1
x1 x 2 t
n n

2 ,n1 + n2 2
2
1

Confident Interval for proportion

Expected Opportunity Loss


N

Ps z

P s (1 P s
n

(P

S1

PS2

i =1

Confident Interval for proportion


( P1 P2 )

z
2

EOL( j ) = (Lij Pi )

p S1 1 p S1 p S2 1 p S 2

n1
n2

df of two population means with unequal


variances
2

S 21
S 2 2
+
n
n 2

1
2
S 21
S 2 2
n

2
+
n1 1
n 2

v =

Sample size for estimating population Z test approximation to the Wilcoxon


mean
Signed Rank Test:
2
n(n + 1)
z
w
n=

4
z=
e
n(n + 1)(2n + 1)
24
Sample size for proportion
Z test approximation to the Wilcoxon Rank2
Sum Test:
z
n = P(1 P)
W1 w11
e
z=
S w1
Hypothesis Tests for the Mean,
z distribution

z=

n
Confident Interval for D
sD
D t

,n 1
n
2

Pooled proportion

X + X2
P= 1
n1 + n 2

Kruskal-Wallis H test statistics


c R 2
12
j
H =
3(n + 1)

n(n + 1) j =1 n j

Spearman Rank Correlation Coefficient


n

6 d i

rs = 1

i =1

n(n 2 1)
Pearson Product
Coefficient

Moment

Correlation

xi y i n x y
i =1

r=
n

xi n x

yi n y

i =1

i =1

Cov( x, y )
SxS y

Two sample test of proportions

pS1 pS 2 ( p1 p2 )

z=

1
1

P1 P +

n1 n2

Expected Monetary Value

Tukey-Kramer Critical Range

Coefficient of Variation
sj
CV ( j ) =
100%
EMV ( j )

Critical Range = Q ,k ,n k

MSW 1
1
+ ,
2 ni n j

where i j

EMV ( j ) = (X ij Pi )
i =1

where s =

[( X

) 2 P( X )

i =1

Chi-square test statistics


( f o f e )2
2
=

fe

Return-to-Risk Ratio

RTRR( j ) =

EMV ( j )
sj

Expected frequency
row column
fe =
n

Standard error of regression model


SSE
SE =
n k 1
F-test for the entire regression model in R- square of regression
multiple regression:
SSR
R2 =
, where SSR is sum square of
MSR
SST
F=
MSE
regression and SST is total sum square.
Testing for the slope in multiple Adjusted R-square of regression

regression:
n 1
2
R
= 1 (1 R 2 )

n k 1
i i
t=

SE i

Confidence interval for the population Paired t test
slope, i

D
D

t
=
,
where:
D
=
i t
SE

i
sD
, n k 1

2
n

i =1

D D

s D = i =1
n 1
The F test statistic for population Variances
n

The Partial F-Test Statistic


(SSRFull SSRRe duced ) / (k Full k Re duced )
F=
MSE Full

S
F = 12
S2

Statistical Tables

You might also like