# However its close approximate with k degrees of

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

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: tions 15 / 26 Two-sample t procedure Suppose we don’t know the populations’ standard deviations σ1 and σ2 . Approximate σ1 −→ s1 and σ2 −→ s2 like before: t= (¯1 − x2 ) − (µ1 − µ2 ) x ¯ 2 s1 s2 +2 n1 n2 Remember that this is an approximation. This t statistic does not exactly follow a t distribution. However, it’s close. Approximate with k degrees of freedom where k is calculated by software, OR k = min(n1 − 1, n2 − 1) K310 Statistical Techniques Inference for Distributions 15 / 26 Two-sample t signiﬁcance test Let’s recap the procedure for a hypothesis test: H0 : µ1 − µ2 = 0 1 State the hypotheses: Ha : µ1 − µ2 = 0 2 Find the test statistic. Let’s break this into a couple steps: 1 2 3 Find the standard error: s(¯1 −x2 ) = x¯ Write the test statistic: t = 2 s1 s2 +2 n1 n2 x1 − x2 ¯ ¯ s(¯1 −x2 ) x¯ Critical region for, e.g., α = 0.05, df = 15: t ∗ = 2.131 −2.131 0 2.131 4 t Find the t statistic relative to the critical region K310 Statistical Techniques Inference for Distributions 16 / 26 Two-sample t signiﬁcance test Let’s recap the procedure for a hypothesis test: H0 : µ1 − µ2 = 0 1 State the hypotheses: Ha : µ1 − µ2 = 0 2 Find the test statistic. Let’s break this into a couple steps: 1 2 3 Write the test statistic: t = 2 s1 s2 +2 n1 n2 Find the standard error: s(¯1 −x2 ) = x¯ x1 − x2 ¯ ¯ s(¯1 −x2 ) x¯ Critical region for, e.g., α = 0.05, df = 15: t ∗ = 2.131 t = 2.5 −2.131 0 2.131 4 t Find the t statistic relative to the critical region Conclusion: t is in the critical region. Reject H0 . K310 Statistical Techniques Inference for Distributions 16 / 26 Two-sample t signiﬁcance test Let’s recap the procedure for a hypothesis test: H0 : µ1 − µ2 = 0 1 State the hypotheses: Ha : µ1 − µ2 = 0 2 Find the test statistic. Let’s break this into a couple steps: 1 2 3 Find the standard error: s(¯1 −x2 ) = x¯ Write the test statistic: t = 2 s1 s2 +2 n1 n2 x1 − x2 ¯ ¯ s(¯1 −x2 ) x¯ Critical region for, e.g., α = 0.05, df = 15: t ∗ = 2.131 t = 1.7 −2.131 0 2.131 4 t Find the t statistic relative to the critical region Conclusion: t is NOT in the critical region. Accept H0 . K310 Statistical Techniques Inference for Dis...
View Full Document

## This note was uploaded on 04/16/2013 for the course MATH-M 310 taught by Professor Palanivelmanoharan during the Spring '13 term at Indiana.

Ask a homework question - tutors are online