Problem: The New York State Lottery

The New York State Lottery Board has commissioned a study of the New York State Lottery. They want a model that allows them to forecast lottery sales. Data is collected on the results of the lottery over 25 weeks. Specifically, they record jackpot size (in thousands of dollars), players’ weekly earnings (in dollars) and the unemployment rate (%) in New York State and lottery sales (in thousands of dollars).

Exhibit 1 : The Available Data

1. Provide an analysis to investigate whether or not there Is there evidence that in weeks when the jackpot is less than $4000, the sales are lower than in other weeks. (10%)
2. Calculate the probability that in a randomly chosen week:

1. The unemployment rate will be above 8%. (2%)
2. Earnings will be above $364 (2%)
3. Earnings will be above $364 and the unemployment rate will be above 8% (3%)
4. Earning will be above $364 given that the unemployment rate is above 8% (3%)

1. Develop the best single variable regression model for sales. Explain your rationale. (10%)
2. Develop the following multiple regression models:

1. Use all available variables
2. Using forwards regression method

Give a brief critique of each of these models and, from all the models you have created in (c) and (d) choose your preferred model. (15%)

1. Consider the model for d(i), which may or may not be the preferred model. Explain the meaning of the Excel output for this model. (10%)
2. For your preferred model, provide an assessment of how well it meets the assumptions of regression. (15%)
3. Give some suggestions as to how you might improve the New York State Lottery Board’s forecasting process. (15%)
4. For the two models in d, calculate a prediction for Sales for the two weeks described in the table below. Comment on how confident you are in each of these predictions. (15%)

Price: $21.98

Solution: The downloadable solution consists of 10 pages, 1198 words and 8 charts.
Deliverable: Word Document