010+Correlation

# 074 df 22 so the hypothesis h0 does not reject but the

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: 1 Restaurant 2 1 1005 918 2 2073 1971 3 873 825 4 1074 999 5 1932 1827 6 1338 1281 7 1449 1302 8 759 678 9 1905 1782 10 693 639 11 2106 2049 12 981 933 9 A. Zhensykbaev Correlation Apply two-tailed test with α = 0.05: H0 = {µ1 − µ2 = 0}, Ha = {µ1 − µ2 = 0}, t-statistic is t= x1 − x2 ¯ ¯ s2 p n1 + s2 p , s2 p (n1 − 1)s2 + (n2 − 1)s2 1 2 = = 273630.6, (n1 − 1) + (n2 − 1) n2 t = 0.384. Reject region RR = {|t| > tα/2 = 2.074 (df = 22). So, the hypothesis H0 does not reject. But the sales of restaurant 1 exceed those of restaurant 2 for every one of the 12 days. It means that we can not apply this test: one of the assumptions of an applicability of t-statistic to test H0 is not true. Let us test the independence using Spearman's test with α = 0.05. Day Restaurant 1 Rank ui Restaurant 2 Rank vi di = ui − vi 1 1005 5 918 4 1 2 2073 11 1971 11 0 3 873 3 825 3 0 4 1074 6 999 6 0 5 1932 10 1827 10 0 6 1338 7 1281 7 0 7 1449 8 1302 8 0 8 759 2 678 2 0 9 1905 9 1782 9 0 10 693 1 639 1...
View Full Document

## This note was uploaded on 02/11/2014 for the course MATH 1390 taught by Professor Christopherstocker during the Spring '13 term at Marquette.

Ask a homework question - tutors are online