Lecture+1+Review+SLR

# Othervariablesthatalsoaffectgpaareleftouteghighschoolp

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: me Y with 100% accuracy for any given X) For example, for any student, given his study time, you cannot precisely predict his GPA, even you have a large sample of data on both. So Y = α + β X won’t be right! Other variables that also affect GPA are left out, e.g. high‐school preparation, IQ, or pure luck on the day of test (one happens to guess half of the exam questions correctly). Alternatively, the dependent variable could have measurement errors, e.g., transcript records could be typed with errors. u is referred to as the error term or the unobservable, is an unobserved random variable that captures all the other variables that affect Y but are omitted from the above equation. When Y is mis‐measured, the measurement error also goes to u , since anything that is part of Y but is not captured by X goes to the error term u. Consider only linear regression Linear regression = linear in parameters ( α ,β ) Examples: Y = α + β lg( Z ) + u , Y = α + β Z 2 + u , Y = α + β e Z + u are all simple linear regressions. Can always rede...
View Full Document

## This document was uploaded on 03/11/2014.

Ask a homework question - tutors are online