Regression and Forecasting Models for Business Applications
STAT 3155

Spring 2015
Baruch College, Statistics & CIS Dept.
Spring 2015
Dr. Yitzchak Sabban
Statistics 3155  Course Outline
Textbooks: FORECASTING, TIME SERIES, AND REGRESSION: AN APPLIED APPROACH
Fourth Edition by Bruce L. Bowerman, Richard T. OConnell and Anne B.
Koehler,
Regression and Forecasting Models for Business Applications
STAT 3155

Spring 2015
STATS 261 SAS LAB FOUR, February 4, 2009
Lab Four: PROC LOGISTIC
Lab Objectives
After todays lab you should be able to:
1.
2.
3.
4.
Use PROC LOGISTIC for multivariate logistic regression.
Interpret output from PROC LOGISTIC.
Understand how to deal with co
Regression and Forecasting Models for Business Applications
STAT 3155

Spring 2015
INFLUENTIAOL OBSERVATION ith
Leverage hi
The leverage of XI of all other observations. h= i observation is usually compared
with the average leverage
 a.
= _L_ + (.X X) . I '
n. gm 4? LWh'Ch equal to the # 0f parameters Including [50
As a rule of thumb
Regression and Forecasting Models for Business Applications
STAT 3155

Spring 2015
Construcng the Normal Probability Plot
From Figure 6. [2(6) we observe a symmcirical plot wilh a pultcmbthm is, :hc pmlern
is linear over a large middle ponion of the plot. However. on each side of [he pica the curve
seems lu atten out. This aucning out s
Regression and Forecasting Models for Business Applications
STAT 3155

Spring 2015
[Ely swan
<3»
INTRODUCTION TO LOGISTIC REGRessION
(OPTIONAL Topic)
multiple leastsquares regression for this type of response variable often leads to predicted
values that are less than zero or greater than one. values that cannot possibly occur.
An alte
Regression and Forecasting Models for Business Applications
STAT 3155

Spring 2015
Slide 1
Copyright 2004
Chapter 4
Probability Distributions
Slide 2
41 Overview
42 Random Variables
43 Binomial Probability Distributions
44 Mean, Variance, and Standard
Binomial Distribution
45 The Poisson Distribution
Copyright