
Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
The methods for selecting random samples and proportions in nonparametric statistical procedures. It outlines the process of generating uniform random numbers and the probability of inclusion based on the sample size and workspace limit. The document also discusses case weighting and its impact on the sampling process.
Typology: Study notes
1 / 1
This page cannot be seen from the preview
Don't miss anything!

1
For each case, a random uniform number in the range 0 to 1 is generated. If it is less than p , the case is included in the sample.
(a) p^
n n n n
1
1
Select a case if its uniform (0,1) number is less than p. If selected, n 1 (^) = n 1 − 1 , and return to (a).
The sampling procedure is as follows: Each time a case is encountered after the limit imposed by the size of the workspace has been reached, the program decides whether to include it in the sample or not at random. The probability that the new cases will enter the sample is equal to the number of cases that can be held in the workspace divided by the number of cases so far encountered. If the program decides to accept a case, it then picks at random one of the cases previously stored in the workspace and drops it from the analysis, replacing it with the new case. Each case has the same probability of being in the sample. If case weighting is used, the nonparametric procedures can use a case more than once. For example, if the weight of a case is 2.3, the program will use that case twice, and may choose at random, with a probability of 0.3, to use it a third time. If sampling is in effect, each of these two or three cases is a candidate for sampling.