Prediction Method

By Rama Krishna Kompella

Multiple Regression

• MR is an intermediate prediction method, allowing:

• 2 or more (usually continuous) IVs

• 1 Continuous DV

• Want IVs relatively uncorrelated

• Want IVs correlated with DV

• Focus is on weights for IVs

Multiple Regression•A regression model specifies a relation between a dependent

variable Y and certain independent variables X1, …,XK.– Here “independence” is not in the sense of random variables; rather, it

means that the value of Y depends on - or is determined by - the Xi variables.)

•A linear model sets Y = 1 + 1X1 + … + kXK + ,

where is the error term. •To use such a model, we need to have data on values of Y

corresponding to values of the Xi's.– selling prices for various house features, past growth values for various

economic conditions

4

When to Use MR?o Standard: Examines how whole set of IVs relates to DV

o Combines all IVs at once to find multiple correlation

o Hierarchical: Examines several sets of IVs based on theory

o Researcher chooses order of variables entered in steps

o Stepwise: Examines IVs most highly correlated with DV

o Computer selects best IVs related to DV

o Conduct any of above in stand-alone MR analysis

o Conduct set of MRs as follow-up to significant Canonical

Correlation

Example

• Suppose we have data on sales of houses in some area. –For each house, we have complete information about its size,

the number of bedrooms, bathrooms, total rooms, the size of the lot, the corresponding property tax, etc., and also the price at which the house was eventually sold.

–Can we use this data to predict the selling price of a house currently on the market?

–The first step is to postulate a model of how the various features of a house determine its selling price.

Example– A linear model would have the following form:

selling price = 0 + 1(sq.ft.) + 2 (no. bedrooms) + 3 (no. bath)

+ 4 (no. acres) + 5 (taxes) + error• In this expression, 1 represents the increase in selling price for each

additional square foot of area: it is the marginal cost of additional area. • 2 and 3 are the marginal costs of additional bedrooms and bathrooms,

and so on. • The intercept 0 could in theory be thought of as the price of a house for

which all the variables specified are zero; of course, no such house could exist, but including 0 gives us more flexibility in picking a model.

Example

– The error reflects the fact that two houses with exactly the same characteristics need not sell for exactly the same price.

• There is always some variability left over, even after we specify the value of a large number variables.

• This variability is captured by an error term, which we will treat as a random variable.

•Regression analysis is a technique for using data to identify relationships among variables and use these relationships to make

predictions.

Levels of advertising

•Determine appropriate levels of advertising and promotion for a particular market segment.

•Consider the problem of managing sales of beer at large college campuses.

– Sales over, say, one semester might be influenced by ads in the college paper, ads on the campus radio station, sponsorship of sports-related events, sponsorship of contests, etc.

•Use data on advertising and promotional expenditures at many different campuses to tell us the marginal value of dollars spent in each category.

•A marketing strategy is designed accordingly. •Set up a model of the following type: sales = 0 + 1(print budget) + 2(radio budget) + 3(sports promo budget) + 4(other promo) + error

9

General Research Questions:

• How do consumers make decisions about the foods that they eat?

• How do these decisions vary across cultures?

More Specific Research Questions:

• What factors influence consumers’ willingness to purchase genetically modified food products?

• Does the influence of these factors vary between U.S. and U.K. consumers?

Descriptive StatisticsU.S. and U.K. Students

US UK N 44 33 Willingness to Purchase

4.86

4.60

General Trust

3.51* 3.22*

Cognitive Trust

5.39* 4.65*

Affective Trust

5.02* 4.41*

Technology

5.17* 4.70*

All student data are on a 7 point scale except general trust which is on a 5 point scale.  †p < .10 *p < .05

Multiple Regression ResultsU.S. and U.K. Students

Dependent Variable: WTP U.S. U.K. Combined General .465† .645† .520* Cognitive -.422† .121 -.167 Affective .892* .271 .649* Technology .179 .000 .000 Country na na -1.09 N 44 31 75 R2 .54 .34 .46

†p < .10 *p < .05

Questions?