Chapter 10--Hypothesis Testing--Categorical Data

We estimate these parameters from the data by x and s

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

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: n this section, we consider a general method of testing for the goodness of fit of a probability model. Example: The frequency distribution of diastolic blood pressure of 14,736 adults ages 30-69 in East Boston, is given below. Group (mmHg) < 50 ≥ 50, < 60 ≥ 60, < 70 ≥ 70, < 80 ≥ 80, < 90 ≥ 90, < 100 ≥ 100, < 110 ≥ 110 Observed Frequency 57 330 2132 4584 4604 2119 659 251 Expected Frequency We wish to test if this data came from a normal distribution. Chapter 10: Hypothesis Testing: Categorical Data Stat 491: Biostatistics Introduction Two-Sample Test for Binomial Proportions McNemar’s Test Estimation of Sample Size and Power R × C Contingency Tables Chi-Square Goodness-of-Fit Test The Kappa Statistic Chi-Square Goodness-of-Fit Test Cont’d... Recall that the normal distribution is indexed by two parameters, µ and σ 2 . ¯ We estimate these parameters from the data by X and S 2 , the sample mean and variance. The expected frequency of the i th class, if the fit is good, is Ei = n × P (Li ≤ X ≤ Ui ) where Li and Ui are the lower and upper class limits of the i th class. In our example, we take L1 = −∞ and U8 = ∞. Chapter 10: Hypothesis Testing: Categorical Data Stat 491: Biostatistics Introduction Two-Sample Test for Binomial Proportions McNemar’s Test Estimation of Sample Size and Power R × C Contingency Tables Chi-Square Goodness-of-Fit Test The Kappa Statistic Chi-Square Goodness-of-Fit Test Cont’d... The agreement between the observed and expected frequencies can be quantified using g X2 = i =1 (Oi − Ei )2 H0 2 ∼ χg −1−k Ei where g is the total number of classes and k is the number of parameters in the model that had to be estimated. Reject H0 if X 2 > χ2 −1−k ,1−α and g 2 p −value = P (χ2 −1−k > Xcomputed ). g Use this test only if no more than (1/5)th of the expected frequencies are < 5 and none of them are < 1. Chapter 10: Hypothesis Testing: Categorical Data Stat 491: Biostatistics Introduction Two-Sample Test for Binomial Proportions McNemar’s Test Estimation of Sample Size and Power R ×...
View Full Document

This note was uploaded on 02/03/2014 for the course STAT 491 taught by Professor Solomonharrar during the Fall '12 term at Montana.

Ask a homework question - tutors are online