Download the Vanderbilt data set (VanderbiltSched.csv), from the course Blackboard page to answer the following questions.

Create a scatter plot of booked elective cases at T-7 vs final case volume.

Create a linear regression model to predict actual case count based on cases scheduled 7 days earlier and show the regression line on the scatter plot.

Evaluate the hypothesis that the average volume of add-on surgeries each day of the week is the same. Here add-on means those added after T-1.

What do you learn by creating a different model for each day of the week?

What do you learn by including dummy variables for each day of the week in a single model?

Develop the best model that you can for predicting daily case volume using the first 45 weeks of data assuming that all relevant decisions must be made by time T-7.

How does your model change if we can make staffing decisions 1 day in advance? Is this better?

Using the final three weeks of data for model validation, how confident are you in the predictions made by the model constructed for question 6?

Using the final three weeks of data for model validation, how confident are you in the predictions made by the model constructed for question 7?

How should your predictions be communicated to inform staff schedule adjustments?