MN 20023 (2013-14)
Week 6 - Bivariate Regression Analysis
Objectives. At the end of these sessions you should:
1. be able to interpret correlation coefficients
2. be able to model relationships using bivariate linear regression and discuss the
strengths and limitations of the method
Introduction
Regression analysis allows us to look at the relationship between a variable we want to predict (e.g. sales) and other variables (e.g. advertising expenditure, sales force size, price and other marketing-mix variables) and thus attempts to give us an explanation of why our sales are behaving as they are. In general terms, regression analysis looks into various aspects (nature, strength, significance etc) of the relationship between a criterion (otherwise called dependent variable) and one or more explanatory variables (otherwise called independent variables). Once identified, such relationships will be useful for Managers to appreciate the effect of explanatory variables having on the criterion variable. Managers can also use them for forecasting among many other uses (e.g. if we know what we will be spending on advertising next month we can forecast what sales will be given other marketing-mix factors remain constant).
Correlation and regression
Correlation is used to measure the strength of association between two variables.
e.g. -how strong is the relationship between our monthly sales and the number of
seconds of television advertising we buy each month?
-how strong is the association between the age of purchasers and their rating of
our product?
-is a larger sales force associated with increased market share?
Regression is used to describe the nature of the association between variables.
e.g. -We expect to sell 50,000 units per month plus an extra 4000 for each minute of
television advertising we buy
Regression can be used to:
1. provide an understanding the relationship between variables...