Fundamentals of market response models Kathrin Gruber Assistant - PowerPoint PPT Presentation

DataCamp Building Response Models in R BUILDING RESPONSE MODELS IN R Fundamentals of market response models Kathrin Gruber Assistant Professor of Econometrics Erasmus University Rotterdam

DataCamp Building Response Models in R Marketing mix

DataCamp Building Response Models in R Market response models Leverage information based on past data to: Adjust product prices. Optimize marketing tactics and strategies. Test the effectiveness of marketing plans. Plan future marketing activities.

DataCamp Building Response Models in R Retail sales str(sales.data) 'data.frame': 124 obs. of 6 variables: $ OBS : int 1 2 3 4 5 6 7 8 9 10 ... $ SALES : num 22.6 22.9 80.6 85.1 81.9 ... $ PRICE : num 1.09 1.27 1.27 1.27 1.27 ... $ DISPLAY : int 0 0 0 0 0 0 0 0 0 0 ... $ COUPON : int 0 0 0 0 0 0 0 0 0 1 ... $ DISPLAYCOUPON: int 0 0 0 0 0 0 0 0 0 0 ... OBS : observation week SALES : volume sales PRICE : average unit price DISPLAY , COUPON , DISPLAYCOUPON : advertising, promotion activities

DataCamp Building Response Models in R Understanding sales mean(sales.data$SALES) 119.7319 min(sales.data$SALES) 11.66749 max(sales.data$SALES) 752.7219

DataCamp Building Response Models in R BUILDING RESPONSE MODELS IN R Let's practice!

DataCamp Building Response Models in R BUILDING RESPONSE MODELS IN R Linear response models Kathrin Gruber Assistant Professor of Econometrics Erasmus University Rotterdam

DataCamp Building Response Models in R Model elements

DataCamp Building Response Models in R Linear response function SALES = f ( PRICE )

DataCamp Building Response Models in R Building a linear response model for sales linear.model <- lm(SALES ~ PRICE , data = sales.data) coef(linear.model) (Intercept) PRICE 274.2486 -134.3097

DataCamp Building Response Models in R Making predictions coef(linear.model)[1] + 0.95 * coef(linear.model)[2] 146.6544 coef(linear.model)[1] + 1.05 * coef(linear.model)[2] 133.2234

DataCamp Building Response Models in R Predictive performance plot(SALES ~ PRICE, data = sales.data) abline(linear.model)

DataCamp Building Response Models in R BUILDING RESPONSE MODELS IN R Let's practice!

DataCamp Building Response Models in R BUILDING RESPONSE MODELS IN R Nonlinear response models Kathrin Gruber Assistant Professor of Econometrics Erasmus University Rotterdam

DataCamp Building Response Models in R Linear response function What happens when PRICE is zero or very large? What about threshold effects?

DataCamp Building Response Models in R Exponential response function Assumes a constant percentage change ( growth rate ).

DataCamp Building Response Models in R Linearizing Nonlinear model: Sales = β ⋆ exp( β ⋆ Price ) 0 1 Linearized model: log( Sales ) = log( β ) + ( β ⋆ Price ) 0 1 log.model <- lm(log(SALES) ~ PRICE, data = sales.data) coef(log.model) (Intercept) PRICE 5.0843983 -0.6622516 percentage change: β ⋆ 100 = (-) 66 % 1

DataCamp Building Response Models in R What's the value added? plot(log(SALES) ~ PRICE, data = sales.data) log.model <- lm(log(SALES) ~ PRICE, data = sales.data) abline(log.model)

DataCamp Building Response Models in R BUILDING RESPONSE MODELS IN R Let's practice!