10

Nonparametric Methods

Embed Size (px)

DESCRIPTION

13. Nonparametric Methods. Introduction So far the underlying probability distribution functions (pdf) are assumed to be known, such as SND, t-distribution, chi-squared distribution Parametric technique (mean, s.d.) Non-parametric techniques - PowerPoint PPT Presentation

Citation preview

Page 1: Nonparametric Methods
Page 2: Nonparametric Methods

13Nonparametric Methods

Page 3: Nonparametric Methods

Introduction

So far the underlying probability distribution functions (pdf) are assumed to be known, such as SND, t-distribution, chi-squared distribution

Parametric technique (mean, s.d.)

Non-parametric techniques Few assumptions about the nature of the underlying pdfNot require the pdf is a SND

Page 4: Nonparametric Methods

13.1 The sign test Investigate the amount of energy expended by patients with the congenital disease cystic fibrosis (CF), and for healthy individuals matched (such as age, sex, height, and weight) to the patients

Rest energy expenditure (kcal/day)

pair CF healthy difference sign

1 1153 996 157 +

2 1132 1080 52 +

3 1165 1182 -17 -

4 1460 1452 8 +

5 1634 1162 472 +

6 1493 1619 -126 -

7 1358 1140 218 +

8 1453 1123 330 +

9 1185 1113 72 +

10 1824 1463 361 +

11 1793 1632 161 +

12 1930 1614 316 +

13 2075 1836 239 +

Table 13.1 Rest energy expenditure for patients with CF and healthy persons.

2 – signs

11 + signs

Page 5: Nonparametric Methods

13.1 The sign test compare the resting energy expenditure (REE) for persons with CF and for healthy individuals (not comfortable in assuming REE or the differences between the measurements are SND) H0 : the median difference is 0H1 : the median difference is not 0It is a two-sided test

(REE)CF – (REE)healthy > 0, < 0, = 0 +, - sign, no information (excluded from the analysis) Under the null hypothesis, we would expect to have approximately equal numbers of + and – signsThat is the probability that a particular difference is + and - are ½Bernoulli random variable with the probability of success p =0.5 Let D = the total number of + signs

Page 6: Nonparametric Methods

13.1 The sign test The mean number of + signs in a sample of size n is np = n/2, and the s.d is (np(1-p))0.5 = (n/4)0.5

If D is either much larger or much smaller than n/2 we would want to reject H0

Evaluate the null hypothesis by considering the test statistic,

If the sample size is large, z+ follows an approximate ND with mean 0 and s.d. 1. This test is called the sign test.

Area to the right and left of 2.50 is p = 2*(0.006) = 0.012 < 0.05 reject the null hypothesis the median difference among pairs is not equal to 0 REE is higher among persons with CF

4/

)2/(

n

nDz

50.24/13

5.611

z

Page 7: Nonparametric Methods

13.1 The sign test If the sample size is small, less than about 20, the test statistic cannot be assumed to have a SND. Therefore, we use the binomial distribution to calculate the probability of observing D positive differences.

Since 0.0112 < 0.05, we would reject the null hypothesis at the 5% level

0112.0

0001.00016.00095.0

)5.0(13

13)5.0(

12

13)5.0()5.0(

11

13

)13()12()11()11(

1313211

DPDPDPDP

Page 8: Nonparametric Methods

Chapter13 p307

13.2 The Wilcoxon Signed-Rank Test

take into account of the magnitude of the pair differences

Page 9: Nonparametric Methods

Chapter13 p307

Page 10: Nonparametric Methods

Chapter13 p311