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: TRUE PROPRORTION of people who receive superior UV protection IS NOT 0.5. PAGE 482: FINDING THE MINIMUM SAMPLE SIZE What to do when the true population proportion is unknown ? What is the meaning of the “minimum sample size”? THE MINIMUM SAMPLE SIZE TO GUARANTEE THE CONFIDENCE LEVEL YOU WANT 8.5 planning a survey of students OVEMBER 21ST SECTION 8.2 TWO POPULATION PROPORTIONS onfidence Interval and Hypothesis Test PROBLEM 8.70 BICYCLE ACCIDENTS REGRESSION rom chapter 10 DON’T WORRY ABOUT onfidence intervals anually calculating bo and b1 inding the MSE or SSE manually AIN IDEA: se date about one variable to predict another O PREDICT, WE CREATE AN EQUATION THAT HAS AN X VARIABLE YHAT is the average response OF ALL SUBJECTS WITH THE GIVEN X Say: model is number of HOURS predict SCORE out of 100 Plug 6 into the x, get yhat = 89 INTERPRETATION: 89 IS THE AVERAGE RESPONSE (SCORE) OF ALL THE MILLIONS AND MILLIONS WHO STUDY 6 HOURS. WHY IS IT THAT REGRESSION PREDICTS THE AVERAGE OF MILLIONS? not the score of 1. ASSUMPTION RESPONSES ARE NORMALLY DISTRIBUTED ASSUMPTION ABOUT ERRORS ERRORS ARE NORMALLY DISTRIBUTED any violation of either assumption will require ANOTHER TYPE OF REGRESSION (non‐linear) OR another type of “blocking” of data to undo nonlinearity of errors TO KNOW IF A MODEL IS GOOD (predictor CAN predict) 1. pvalue for the model is less than alpha (say alpha = 0.05) 2....
View Full Document

This document was uploaded on 01/28/2014.

Ask a homework question - tutors are online