This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: MS&E 211 Fall 2007 Linear and Nonlinear Optimization Nov 19, 2007 Prof. Yinyu Ye Homework Assignment 6: Due 3:15pm Tuesday, Dec 4 There is a homework collecting box outside of professor Yes office (Terman 316) for you to submit your homework. No late homework accepted! Problem 1: Logistic Regression Microsoft would like to predict if a Stanford graduate will make a good employee based on his graduating GPA. Their human resources department has a tradition of keeping files about all their employees for years even after firing them. Your are asked to use this information to build a tool that will predict the likelihood that a Stanford graduate will make a good employee based solely on the candidates GPA. To do so, you will use a popular method for binary classification called logistic regression. Let c i be the GPA of candidate i and let y i represent the outcome of an evaluation period (i.e. y i = 0 means the candidate was fired after the evaluation period, y i = 1 means that he became full-time employee). A logistic regression model assumes that: P ( y i = 1 | c i ,x ) = f ( c i x 1 + x 2 ) , P ( y i = 0 | c i ,x ) = 1...
View Full Document
This note was uploaded on 06/16/2010 for the course MS&E 211 taught by Professor Yinyuye during the Fall '07 term at Stanford.
- Fall '07