Economics for Managers by Paul Farnham Chapter 4: Techniques for Understanding Consumer Demand and Behavior 4.1
Getting Information About Consumer Behavior Expert opinion Consumer surveys Test marketing and price experiments Analyses of census and other historical data Regression analysis 4.2
Managerial Rule of Thumb: Analyzing Consumer Behavior Managers must consider 1. Whether the participating groups are truly representative of the larger population 2. Whether the answers given in these formats represent actual market behavior 3. How to isolate the effect of different variables influencing demand 4.3
Economic Approaches to Consumer Behavior Multiple regression analysis Estimates the relationship between a dependent variable and an independent variable holding constant the effects of all other variables Can be done with spreadsheet software 4.4
Economic Approaches to Consumer Behavior Simple regression analysis Cross-sectional data Time series data Panel data Provides an equation that best fits the data 4.5
Hypothetical Demand for Oranges Figure 4.1 Pr rice (c cents/l b.) 150 100 50 0 50 100 150 Quantity (lb.) 4.6
Quantitative Measure Linear demand relationship can be shown as this equation: Q = a - bp where Q = the quantity demanded a = vertical intercept b = the slope of the line = ΔQ / ΔP P = price 4.7
Simple Regression Analysis Form of regression analysis that analyzes the relationship between one dependent and one independent variable Can be done easily using the Microsoft Excel Regression feature 4.8
Significance of the Coefficients i Two important issues: Hypothesis testing for the significance of the estimated coefficients The goodness of fit of the entire estimating equation 4.9
Standard Error Small standard error means coefficient would not vary much among regressions T-test is based on ratio of the size of coefficient to its standard error Large values of t-statistic show significant results 4.10
Significance of the Coefficients i Confidence intervals Coefficient of determination, R 2 F-statistic 4.11
Multiple Regression Analysis Q = a bp + cadv where Q = the quantity demanded d d a = vertical intercept b = the slope of the line = ΔQ Q/ ΔPP P = price c = coefficient of advertising variable ADV = advertising expenditure 4.12
Multiple Regression Analysis Figure 4.4 Fit for Advertising Variable Y Predicted Y Qu uantity 200 100 0 2.00 4.00 6.00 8.00 Advertising 4.13
Other Functional Forms Log-linear demand function Can be linear by taking the logarithms of all variables in the following equation: Q x = (a) (P b x ) (ADV c ) Can also be called constant-elasticity demand d function Price and advertising elasticities can be read directly from statistical ti ti results 4.14
Log-Linear Demand Curve Figure 4.5 x x x 0 x x x x x x Demand x x Quantity 4.15
Demand Estimation Issues Examples are not suitable for complex analyses, but can provide a starting point Regression analysis can be influenced by data availability and underlying economic theory All analysis is influenced by sample of data used 4.16
Managerial Rule of Thumb: Using Multiple l Regression Manager must decide which variables to include in an analysis Problems can arise if relevant variables are excluded or irrelevant variables are included 4.17
Managerial Rule of Thumb: Using Multiple l Regression Choice of variables comes from Economic theory Real-world experience The problem under consideration Common o sense 4.18
Managerial Rule of Thumb: Using Empirical Consumer Demand Studies Empirical consumer demand studies show types of data available for analyzing the demand for different products Many data sources may not be widely ypublicized Some studies use previous analyses 4.19
Managerial Rule of Thumb: Using Consumer Market Data Statistical and econometric models can assist in developing competitive strategies t by indicating the importance of characteristics influencing demand for different products Studies also show what tradeoffs consumers may be willing to make 4.20
Summary of Key Terms Coefficient i of determination ti Cross-sectional data Direct consumer surveys Expert opinion Multiple regression analysis Panel data 4.21
Summary of Key Terms Pi Price experiments Simple regression analysis Standard error Targeted marketing Test marketing Time-series data T-test 4.22