Suppose we have many variables in a data set for an ice cream company. We wish to predict the sales (in dollars), so the y-variable for our model will be sales. We have these additional x-variables: • temperature • humidity • precipitation • advertising budget • discount rate Suppose we would like to crate a simple linear regression model that predicts the sales based on the variable that is the most strongly correlated with sales. How should we create the desired model? OWe should create a correlogram and look for the largest dot in the Sales row. The variable associated with this largest dot should be used to create a linear regression model. We should first create a "full" model that contains all 5 x-variables. Then we should omit all of the variables who have a high p-value, and make a simplified model consisting of exactly the variables whose slopes had a small p-value in the "full" model. To most accurately answer this question, include all 5 x-variables (temperature, humidity, precipitation, advertising budget, and discount rate). We should create a "full" model with all 5 x-variables, and then use the step() function in R to calculate a reduced model.
Suppose we have many variables in a data set for an ice cream company. We wish to predict the sales (in dollars), so the y-variable for our model will be sales. We have these additional x-variables: • temperature • humidity • precipitation • advertising budget • discount rate Suppose we would like to crate a simple linear regression model that predicts the sales based on the variable that is the most strongly correlated with sales. How should we create the desired model? OWe should create a correlogram and look for the largest dot in the Sales row. The variable associated with this largest dot should be used to create a linear regression model. We should first create a "full" model that contains all 5 x-variables. Then we should omit all of the variables who have a high p-value, and make a simplified model consisting of exactly the variables whose slopes had a small p-value in the "full" model. To most accurately answer this question, include all 5 x-variables (temperature, humidity, precipitation, advertising budget, and discount rate). We should create a "full" model with all 5 x-variables, and then use the step() function in R to calculate a reduced model.
Related questions
Question
Transcribed Image Text:Suppose we have many variables in a data set for an ice cream company. We wish to
predict the sales (in dollars), so the y-variable for our model will be sales.
We have these additional x-variables:
• temperature
• humidity
• precipitation
• advertising budget
• discount rate
Suppose we would like to crate a simple linear regression model that predicts the
sales based on the variable that is the most strongly correlated with sales.
How should we create the desired model?
We should create a correlogram and look for the largest dot in the Sales row.
The variable associated with this largest dot should be used to create a linear
regression model.
OWe should first create a "full" model that contains all 5 x-variables. Then we
should omit all of the variables who have a high p-value, and make a simplified
model consisting of exactly the variables whose slopes had a small p-value in the
"full" model.
To most accurately answer this question, include all 5 x-variables (temperature,
humidity, precipitation, advertising budget, and discount rate).
We should create a "full" model with all 5 x-variables, and then use the step()
function in R to calculate a reduced model.
Expert Solution
This question has been solved!
Explore an expertly crafted, step-by-step solution for a thorough understanding of key concepts.
This is a popular solution!
Trending now
This is a popular solution!
Step by step
Solved in 2 steps
