This preview shows pages 1–5. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Leveraged Outliers Dependent Errors and Time Series Outline Leveraged Outliers Dependent Errors and Time Series 1 / 15 ISOM 2500 Lect 23: Leveraged Outliers; Dependent Errors and Time Series Leveraged Outliers Dependent Errors and Time Series How Much to Bid? • A contractor is bidding on a project to construct an 875 squarefoot addition to a home. • If he bids too low, he loses money on the project. • If he bids too high, he does not get the job. • The contractor has kept data that record the costs for 30 similar project: • All but one of his previous projects are smaller than 875 square feet. • His one project at 900 square feet is an outlier. 2 / 15 ISOM 2500 Lect 23: Leveraged Outliers; Dependent Errors and Time Series Leveraged Outliers Dependent Errors and Time Series Keep or Remove the Outlier? • If keep the outlier, then the fitted equation is Estimated Cost ($) = 5887 . 74 + 27 . 44 * Size (sq.ft.) • If remove the outlier, then the fitted equation becomes Estimated Cost ($) = 1558 . 17 + 44 . 74 * Size (sq.ft.) 3 / 15 ISOM 2500 Lect 23: Leveraged Outliers; Dependent Errors and Time Series Leveraged Outliers Dependent Errors and Time Series Leveraged Observations • Say that an observation is leveraged in regression if it has a small or large value of the explanatory variable. • The 900 square feet outlier is a leveraged observation as it has a large value of the explanatory variable (actually much larger than the remaining) • A leveraged observation has the ability to pull the regression line in its direction, like in the bid case 4 / 15 ISOM 2500 Lect 23: Leveraged Outliers; Dependent Errors and Time Series Leveraged Outliers...
View
Full
Document
This note was uploaded on 12/20/2011 for the course ACCT/MGMT 2010 taught by Professor A during the Spring '11 term at HKUST.
 Spring '11
 a

Click to edit the document details