formulat sheet - (1-ˆ p 2 n 2 N(0 1 n i p i> 5 d 6 = 0 n...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Test Null Hypothesis (Param. = Value) Param. Statistic Std. Error Test Statistic ± Statistic - Parameter Std.Error ² Assumptions 1-sample t-test H 0 : μ = μ 0 μ 0 ¯ x s n t ( df ) n < 15 and Normal (or) 15 < n < 40 and No Outliers (or) df = n - 1 n > 40 Matched Pair t-test H 0 : d = d 0 d 0 ¯ d s n t ( df ) n < 15 and Normal (or) df = n - 1 15 < n < 40 and No Outliers (or) n > 40 2-sample t-test, Unpooled H 0 : μ 1 - μ 2 = d 0 d 0 ¯ x 1 - ¯ x 2 q s 2 1 n 1 + s 2 2 n 2 t ( df ) Normality (or) n 1 + n 2 > 40 df = min( n 1 ,n 2 ) - 1 σ 1 6 = σ 2 and unknown 2-sample t-test, Pooled H 0 : μ 1 - μ 2 = d 0 d 0 ¯ x 1 - ¯ x 2 s p q 1 n 1 + 1 n 2 t ( df ) Normality (or) n 1 + n 2 > 40 s p = q ( n 1 - 1) s 2 1 +( n 2 - 1) s 2 2 n 1 + n 2 - 2 df = n 1 + n 2 - 1 σ 1 = σ 2 and unknown 1-proportion z-test H 0 : p = p 0 p 0 ˆ p q p 0 (1 - p 0 ) n N (0 , 1) np 0 > 10 n (1 - p 0 ) > 10 2-prop. z-test, Pooled H 0 : p 1 - p 2 = 0 0 ˆ p 1 - ˆ p 2 r ˆ p (1 - ˆ p ) ³ 1 n 1 + 1 n 2 ´ N (0 , 1) n i p i > 5 ˆ p = n 1 p 1 + n 2 p 2 n 1 + n 2 n i (1 - p i ) > 5 2-prop. z-test, Unpooled H 0 : p 1 - p 2 = d 0 d 0 ˆ p 1 - ˆ p 2 q ˆ p 1 (1 - ˆ p 1 ) n 1 + ˆ p 2
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: (1-ˆ p 2 ) n 2 N (0 , 1) n i p i > 5 d 6 = 0 n i (1-p i ) > 5 Chi-Square Test A NOT assoc. with B χ 2 = ∑∑ ( O ij-E ij ) 2 E ij E ij > 1 df = ( r-1) × ( c-1) E ij < 5 for ATMOST 20% cells Linear Regression H : β i = μ i μ i b i se b i t ( df ) df = N-C Linear Regression H : β 1 = ...β C-1 = 0 F = MS R MS E df n = C-1 ,df d = N-C ANOVA H : μ 1 = ... = μ G F = MS B MS W df n = G-1 ,df d = N-G...
View Full Document

This note was uploaded on 08/31/2010 for the course MANAGEMENT MGCR 271 taught by Professor Vaidyanathan during the Summer '10 term at McGill.

Ask a homework question - tutors are online