R Tutorial: Linear response models

Want to learn more? Take the full course at https://learn.datacamp.com/courses/building-response-models-in-r at your own pace. More than a video, you'll learn hands-on coding & quickly apply skills to your daily work. --- Now as we are familiar with the data we move on to constructing a simple response model for sales. Market response models aim to relate the desired outcome from the target market to the marketing tools. Customers respond to advertising and promotion through changes in purchase behavior. Sales are the most direct measures of the outcome of this activities. Therefore, in a sales response model, SALES is defining the response variable we are trying to explain. And those variables that are providing the explanation, the so-called predictor variables, commonly consist of the advertising and promotion activities. For the moment we say that there is only one predictor variable, PRICE, that was thought to influence SALES. Once the response and the predictor variables are selected, one has to decide on the form of the relationship between the variables in the model, the so-called response function. The simplest form to assume, for describing the sales-price relation, is to assume linearity. More specifically, we assume SALES decrease by a constant amount when PRICES increase by a constant amount. The corresponding linear response model consists of two parts: the intercept beta zero, which reflects the average level of sales if price is zero, and the slope beta one, which describes how the average level of sales changes with a unit change in price. A linear response model is, in fact, a simple linear model and can be estimated by using the linear model function. We describe the relationship between the SALES response and the PRICE predictor via a symbolic formula argument and store the estimated results in an object called linear-dot-model. The estimated intercept and slope coefficients can be obtained from the linear-dot-model object by using the