Necessary Conditions for Statistical Decision Making-ECO6416

Necessary Conditions for Statistical Decision...

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

View Full Document Right Arrow Icon
Necessary Conditions for Statistical Decision Making Introduction to Inferential Data Analysis Necessary Conditions: Do not just learn formulas and number-crunching. Learn about the conditions under which statistical testing procedures apply. The following conditions are common to almost all statistical tests: 1. Any undetected outliers may have major impact and may influence the results of almost all statistical estimation and testing procedures. 2. Homogeneous population. That is, there is not more than one mode. Perform Test for Homogeneity of a Population 3. The sample must be random. Perform Test for Randomness . 4. In addition to the Homogeneity requirement, each population has a normal distribution. Perform the Lilliefors' Test for Normality . 5. Homogeneity of variances. Variation in each population is almost the same as in the other(s). Perform The Bartlett's Test . For two populations use the F-test. For 3 or more populations, there is a practical rule known as the"Rule of 2". In this rule, one divides the highest variance of a sample by the
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: lowest variance of the other sample. Given that the sample sizes are almost the same, and the value of this division is less than 2, then the variations of the populations are almost the same. Notice: This important condition in analysis of variance (ANOVA and the t-test for mean differences) is commonly tested by the Levene test or its modified test known as the Brown-Forsythe test. Interestingly, both tests rely on the homogeneity of variances condition! These conditions are crucial, not for the method of computation, but for the testing using the resultant statistic. Otherwise, we can do ANOVA and regression without any assumptions, and the numbers come out the same. Simple computations give us least-square fits, partitions of variance, regression coefficients, and so on. We do need the above conditions when test of hypotheses are our main concern....
View Full Document

This note was uploaded on 10/04/2011 for the course ECO 6416 taught by Professor Staff during the Spring '08 term at University of Central Florida.

Ask a homework question - tutors are online