Scribd
Upload
141 views4 pages

Confidence Intervals and Tests: List of Tables

This document discusses confidence intervals and hypothesis tests for proportions, means, and regression parameters. It provides tables that describe the relevant statistical methods, including the sample, parameter, hypotheses, test statistic, degrees of freedom, and relevant Excel commands for confidence intervals and one-sample, two-sample, and regression tests.

Uploaded by

Asbjørn May
Copyright
© Attribution Non-Commercial (BY-NC)
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
141 views4 pages

Confidence Intervals and Tests: List of Tables

This document discusses confidence intervals and hypothesis tests for proportions, means, and regression parameters. It provides tables that describe the relevant statistical methods, including the sample, parameter, hypotheses, test statistic, degrees of freedom, and relevant Excel commands for confidence intervals and one-sample, two-sample, and regression tests.

Uploaded by

Asbjørn May
Copyright
© Attribution Non-Commercial (BY-NC)
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/ 4

Condence intervals and tests

Methods III Statistics and Quantitative Methods Copenhagen Business School Fall, 2011

List of Tables
1 2 3 Condence intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hypothesis tests (proportions and means) . . . . . . . . . . . . . . . . . Hypothesis tests (contingency tables and estimated regression parameters) 2 3 4

Table 1: Condence intervals


Description Proportions One proportion Dierence Means One mean Dierence - uneq. pop. var. - eq. pop. var. 1 2 1 2 x1 x2 x1 x2 sex1 2 = x sex1 2 = s x s= Regression analysis Single parameter
s2 1 n1

Parameter

Sample

Standard error sep =


p(1) p n p1 (11 ) p n1 p2 (1p2 ) n2

Condence interval p z1 sep 2 + p1 p2 z1 sep1 2 p 2 x t sex 2

Degrees of freedom

Relevant Excel 2007 commands AVERAGE, COUNT, NORMSINV(1 ) 2 AVERAGE, COUNT, NORMSINV(1 ) 2

p p1 p2

p p1 p2

sep1 2 = p sex =
s n

n1

AVERAGE, STDEV, COUNT, TINV(;deg. of free.)

s2 2 n2 1 n2

x1 x2 t sex1 2 x 2 x1 x2 t sex1 2 x 2

Welch-Satterthwaite AVERAGE, STDEV, COUNT, TINV(;deg. of free.) Approx. n1 + n2 2 n1 + n2 2 AVERAGE, STDEV, COUNT, TINV(;deg. of free.)

1 n1

(n1 1)s2 +(n2 1)s2 1 2 n1 +n2 2

bk

sebk

bk t sebk 2

n No. of param.

Regression in Data Analysis

: Signicance level, probability in NORMSINV is the left tail probability while probability in TINV is the sum of the tail probabilities.

Table 2: Hypothesis tests (proportions and means)


Description Proportions One proportion - one sided - one sided - two sided Dierence - one sided - one sided - two sided Means One mean - if t > 0: one sided one sided two sided - if t < 0: one sided one sided two sided Dierence - unequal pop. variances - equal pop. variances 1 2 1 2 x1 x2 x1 x2 H0 : 1 2 = 0 H0 : 1 2 = 0 se0 = se0 = s s= - if t > 0: one sided Ha : one sided Ha : two sided Ha : - if t < 0: one sided Ha : one sided Ha : two sided Ha : : Another approach: Use Data Analysis or the function 1 2 < 0 1 2 > 0 1 2 = 0 1 2 < 0 1 2 > 0 1 2 = 0 TTEST
s2 1 n1

Parameter

Sample

Hypotheses

Standard error under H0 se0 =


p0 (1p0 ) n

Test statistic z=
pp0 se0

Degrees of freedom

Relevant Excel 2007 commands AVERAGE, COUNT NORMSDIST(z) 1 NORMSDIST(z) 2 NORMSDIST(-ABS(z)) AVERAGE, COUNT NORMSDIST(z) 1 NORMSDIST(z) 2 NORMSDIST(-ABS(z))

H0 : Ha : Ha : Ha :

p = p0 p < p0 p > p0 p = p0

p1 p2

p1 p2

H0 : p1 p2 = 0 Ha : p1 p2 < 0 Ha : p1 p2 > 0 Ha : p1 p2 = 0

se0 =

p(1 p)

1 n1

1 n2

z=

p1 2 p se0

p: Pooled estimate

H0 : = 0 Ha : < 0 Ha : > 0 Ha : = 0 Ha : < 0 Ha : > 0 Ha : = 0

se0 =

s n

t=

x0 se0

n1

AVERAGE, STDEV, COUNT 1 TDIST(t; deg. of freedom; one tail) TDIST(t; deg. of freedom; one tail) TDIST(t; deg. of freedom; two tails) TDIST(ABS(t); deg. of freedom; one tail) 1 TDIST(ABS(t); deg. of freedom; one tail) TDIST(ABS(t); deg. of freedom; two tails)

s2 2 n2 1 n2

t= t=

x1 2 x se0 x1 2 x se0

Welch-Satterthwaite Approx. n1 + n2 2 n 1 + n2 2

AVERAGE, STDEV, COUNT AVERAGE, STDEV, COUNT

1 n1

(n1 1)s2 +(n2 1)s2 2 1 n1 +n2 2

1 TDIST(t; deg. of freedom; one tail) TDIST(t; deg. of freedom; one tail) TDIST(t; deg. of freedom; two tails) TDIST(ABS(t); deg. of freedom; one tail) 1 TDIST(ABS(t); deg. of freedom; one tail) TDIST(ABS(t); deg. of freedom; two tails)

Table 3: Hypothesis tests (contingency tables and estimated regression parameters)


Description Contingency tables Chi-squared Parameter Sample Hypotheses Standard error under H0 Test statistic X 2 = All cells (ObservedExpected) Expected Expected = Row totalColumn total Grand total
2

Degrees of freedom (Rows 1)(Cols 1)

Relevant Excel 2007 commands CHIDIST(X 2 ;deg. of freedom)

H0 : Independence Ha : Association H0 : Independence Ha : Association

Fishers exact test

p-value = (R1 !R2 !R!m !)(C1 !C2 !Cn !) N ij aij ! R: Row total C: Column total N : Grand total a: Cell count m: Number of rows n: Number of columns

FACT, PRODUCT

Regression analysis Single parameter - if t > 0: one sided one sided two sided - if t < 0: one sided one sided two sided All parameters

bk

H0 : k = 0 Ha : k < 0 Ha : k > 0 Ha : k = 0 Ha : k < 0 Ha : k > 0 Ha : k = 0

se0

t=

bk se0

n No. of param.

Regression in Data Analysis 1 TDIST(t; deg. of freedom; one tail) TDIST(t; deg. of freedom; one tail) TDIST(t; deg. of freedom; two tails) TDIST(ABS(t); deg. of freedom; one tail) 1 TDIST(ABS(t); deg. of freedom; one tail) TDIST(ABS(t); deg. of freedom; two tails)

1 , . . . , k

b1 , . . . , bk

H0 : 1 = = k = 0

F =

Mean square for regression M eansquareerror

df1 = No. of exp. var. df2 = No. of param.

Regression in Data Analysis

Ha : one = 0 Another approach: Use the function CHITEST

You might also like