Inferences on p class

Baker department of statistics g university of south

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: rtment of Statistics G. University of South Carolina; Slide 7 ˆ Sampling Distribution for p Sampling ˆ The sampling distribution of p based on large n is approximately normal. Rule of Thumb: If np > 5 and n(1-p) > 5 (preferably both > 10), then the distribution of Y and hence Y/n can be approximated with a normal distribution. G. baker, Department of Statistics G. University of South Carolina; Slide 8 Sampling Distribution for ˆ p To completely define the normal distribution of ˆ p We need the mean (expected value) and variance. G. baker, Department of Statistics G. University of South Carolina; Slide 9 ˆ Sampling Distribution of p Sampling 1-α p − zα / 2 p (1 − p ) n p p + zα / 2 p (1 − p ) n p(1 − p ) n ˆ So, at most, p will be away from p, (1-α)100% of the time. We call this (1-α)100% the level of confidence. zα / 2 G. baker, Department of Statistics G. University of South Carolina; Slide 10 Repeated Sampling of Size n 95% 95% of the time our estimate will be within p (1 − p ) 1.96 n of the truth. G. baker, Department of Statistics G. University of South Carolina; Slide 11 Standard Error We don’t know the value of p, so we will We know use p use ˆ ˆ When we use p , we have an estimate of the standard deviation for the sampling ˆ distribution of p . distribution We call this estimate the standard error. ˆ ˆ p (1 − p) n G. baker, Department of Statistics G. University of South Carolina; Slide 12 Confidence Interval for p ˆ (1 − α )100%CI for p : p ± zα / 2 ˆ ˆ p(1 − p ) n Example: ˆ ˆ p (1 − p ) ˆ 95%CI for p : p ± 1.96 n G. baker, Department of Statistics G. University of South Carolina; Slide 13 Confidence Interval Based on Confidence Normal Distribution Normal Pt Est ± ( Z Value) (Standard Error) ˆ p ± zα / 2 ˆ ˆ p (1 − p ) n Standard error is our estimate of the standard deviation for the distribution of the point G. baker, Department of Statistics G. University of South Carolina; Slide 14 estimate. Confidence Interval Estimation Airplanes approaching a runway for a landing are required to stay within...
View Full Document

This note was uploaded on 01/12/2014 for the course STA 509 taught by Professor Wang during the Fall '13 term at South Carolina.

Ask a homework question - tutors are online