SIMPLE lINEAR rEGRESSION英文版完整六西格玛方案.ppt

Simple Linear Regression Learning Objectives Overview of Regression Common Regression Mistakes Common Regression Mistakes Linear Regression Exercise End of TopicWhat question do you have? Conclusion: On the average, loss of 7.281 mils in groove depth for every 1000 miles of driving. Regression Diagnostics Verifying Outliers and Influential Observations Common reasons for removing outliers and influential observations from dataset: Measurement error Data entry error Violation of a known physical law Note: No regression equation that contains an obviously bogus value can be considered valid! Simple Linear Regression Applications of Regression Model Prediction System Explanation Parameter Estimation Simple Linear Regression Prediction of Future Observations Common use of regression model is to predict value of Y given some fixed value of x. can be used for 2 inference problems: To predict a single or individual value of Y, denoted by yi To estimate the fixed unknown mean of Y E.g, for the car tire problem: To an individual who purchases a car tire, regression model helps to predict groove depth of his new tire after certain number of miles. To a dealer who purchases large quantities of the tires, model helps to estimate mean groove depth of the population of tires after certain number of miles. Simple Linear Regression Prediction of Future Observations To reflect precision in estimating the mean value, we report a confidence interval for the estimate. To reflect precision in predicting individual value, we report a prediction interval for the estimate. The prediction interval is wider than the confidence interval because there is more uncertainty associated with predicting individual rather than average values. Simple Linear Regression Example 10 – Confidence and Prediction Intervals Find the confidence and prediction intervals for the car tire problem of Example 1. (Data found in SLR.MTW) Minitab: Stat ? Regression ?