• ## Summary

Search form
• 00:00

[CUSTOMER ANALYTICS, Raghu Iyengar, Regression Analysis,Wharton University of Pennsylvania ONLINE]

• 00:04

RAGHU IYENGAR: Welcome to Customer Analytics.So as Pete mentioned, there are broadlytwo ways in which you can think about quantifying data.One is making predictions one period ahead.The other is making predictions more than two periods ahead.So in this module, we'll talk about the first one,making predictions one period ahead.How do we do that?It's done through regression analysis.So what are we going to do in this module is

• 00:25

RAGHU IYENGAR [continued]: talk about a simple example, show how regressioncan be done, show what predictions it can make,and then we'll take it off with Pete, who will talkmore about two periods ahead.So let's start with regression analysis.What's regression all about?It's about quantifying the relationshipbetween two or more variables.Let's take a simple example.Suppose you're looking at demand or data of people purchasing.

• 00:47

RAGHU IYENGAR [continued]: And you know how prices will change.What we'd like to do is to think about howyou can start thinking about how price is changing demand.In other words, put some numbers behind it.Let's look at some jargon of regression.What we are trying to do is to explain a dependent variable--in this case sales or demand--

• 01:08

RAGHU IYENGAR [continued]: as a function of independent variables-- in this case price.In other words, all we're trying to do in regressionis try to make predictions of what would bethe demand at different prices.Regression is a technique that uses simple linear additivemodel to make these kinds of predictions.It'll become clear by taking a simple example.

• 01:30

RAGHU IYENGAR [continued]: Let's imagine this is the demand data for a particular firmat different prices.What this firm is trying to do isto try and understand how their prices might change demand.So they ended up changing the prices.And they observe the demand.The very first thing we should dowhen you start thinking about quantifying the relationshipis just plot the data.

• 01:51

RAGHU IYENGAR [continued]: So let's plot it.Here's what the plot looks like.What do we see here?On the horizontal axis, we have price.On the vertical axis, we have sales.And what we see here, which is what intuitively you'dexpect to see, is that as prices go up, sales come down.On the one hand it's intuitive.It makes a lot of sense.

• 02:12

RAGHU IYENGAR [continued]: And this is what you would call a demand curve,prices going up, sales coming down.Where does regression come in?Regression comes in to give some hard numbers.You can eyeball it and see that as you increase price,sales do come down.But we would like to see it specifically by how much.In other words, we'd like to answer the following question.

• 02:32

RAGHU IYENGAR [continued]: If I increase price by $1, how much the sales come down?That's where regression comes in.What does regression do?It tries to fit a straight line to the datathat we see here and tries to put formal numbersbehind this demand curve.Broadly what we talk about in a simple exampleis demand analysis. • 02:53 RAGHU IYENGAR [continued]: This is a specific example for regression.You can think about doing it for many other types of data.What we are doing here is sales as a function of price.You can think about sales as a function of advertising.You can think about a variety of different variablesthat you'd like to see if they're connected together.So the simplest form of regression analysisthat we can do here is sales, which • 03:14 RAGHU IYENGAR [continued]: is our dependent variable, is a function of price, whichis our independent variable.So if you look on the left-hand side, we have sales.On the right-hand side, we have price.The coefficient b, which is in front of price,basically measures price sensitivity.In the next slide, I'm going to show youhow getting an understanding of what b iswill basically help us understand if I increase price • 03:37 RAGHU IYENGAR [continued]: by$1, how much would the sales come down?Now this equation that we see hereis a form of a general regression example, where youcan think about sales as being representedby y and price being represented by x.So in the equation below, what I've shown

• 03:57

RAGHU IYENGAR [continued]: you is a general form where you canthink about putting in many different y's that youwill care about.For example, if you're in a company thatlooks at advertising and sales, in your example,y would be sales.That's what you're trying to predict.That's a dependent variable.And x would be advertising.That's your independent variable.In this example that we're showing you here,

• 04:19

RAGHU IYENGAR [continued]: y is sales as well.x is price.Once you run the regression, what you would see hereis a regression or predicted line.That's the line that you see on the graph there.In the line you also see a regression equation,which basically tells you how your sales and prices areconnected together.And you also see something called an R square.

• 04:42

RAGHU IYENGAR [continued]: Let me first give an intuition of what R square is.R square basically tells you how good is the regression line.The more scatter that you see from the straight line, Rsquare will be small.In other words, the straight lineis not able to capture all the variations.The more the lines are closer to the straight line,you will see that R squared is quite high, closer to 1.

• 05:03

RAGHU IYENGAR [continued]: In other words, the regression is doing quite a good job.Once you determine that the regression is doing a good job,typically R square about 70% to 80%,then you can go ahead and start using this regressionfor making predictions.And that's what we will do next.

### Video Info

Series Name: Customer Analytics

Episode: 11

Publisher: Wharton

Publication Year: 2016

Video Type:Tutorial

### Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:

Keywords:

## Abstract

Raghu Iyengar, Professor of Marketing at Wharton University, introduces regression analysis and discusses the demand curve in part 1 of regression analysis of the predictive analytics module.