We can expand upon the analysis above by exploring the spring and fall seasons.

In the plot above, we can graph overall total vs$.$ temperature for spring and fall on the same axes, where we

create a filter to show only spring and fall. The model with an interaction term we made for question $6$

indicates that it is appropriate to create parallel trend lines for spring and fall. We know that Tableau will not

create the parallel trend lines for us$,$ however a multiple linear regression model using R will be able to create

parallel trend lines.

There are a few ways to create these parallel trend lines in R$.$ The method that we have been using in reading

assignments and labs this semester is to first create a subset of the data. The code below can be used to create a

new data set named SpringFallData, where the only seasons that are kept in this new data set are spring and

fall. You will need to run this code in R$,$ and use the SpringFallData data set in question $7.$

SpringFallData $<-$ subset$($BikesData$,$ Season $\%$in$\%$ c$("$spring$",$ "fall"$))$

Create a model that uses only data from the spring and fall. Create a model where the dependent variable is

OverallTotal, and the independent variables are TempF and Season. Do not include an interaction term.

Using the data from only spring and fall, when we control for season, whenever the temperature increases $5$

degrees Fahrenheit, the predicted number of overall bike rentals increases by $\_\_\_\_\_\_\_.$

Using the data from only spring and fall, when we compare days with the same temperature, the predicted

number of overall bike rentals is $\_\_\_\_\_$ higher in the fall than in the spring.