This preview shows pages 1–3. Sign up to view the full content.
Linear Regression
Data Mining
Prof. Dawn Woodard
School of ORIE
Cornell University
1
Outline
1
Announcements
2
The Regression Task
3
Simple Linear Regression
2
Announcements
Questions?
Frequentist intervals for naive Bayes
Finish crossvalidation
4
The Regression Task
So far we have considered the
classi±cation task
In this task the goal is to predict the value of a
categorical
outcome
In order to do this we have training data that includes the
value of both
predictors and outcome
Another type of supervised learning is prediction of the
value of a
continuous outcome
We will still have training data that includes the value of
both
predictors and outcome
6
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentTV Ads Data
Amount of spending by 21 companies on
TV ads
(“SPEND”)
For each company, we also know the number of
retained
impressions
per week in millions (“MILIMP”)
Based on a survey of 4,000 adults
8
TV Ads Data
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Ad Spending (Millions $)
This is the end of the preview. Sign up
to
access the rest of the document.
 '08
 DAWN

Click to edit the document details