Coffee Exports Country Venezuela Sri_Lanka Nigeria Philippines Togo Paraguay Sierra_Leone Ghana Jamaica Angola Zimbabwe Zambia Dominican_Republic...

There are three data sets for you to analyze.

Data Set 1 (Download Excel File "Coffee Exports")

Data set 1 has coffee exports (thousands of 60-kilo bags) for December 2000 and total exports for 2001 for 42 coffee exporting countries excluding the three largest exporters (Brazil, Colombia, and Vietnam). A scatterplot is shown below.

Commodity traders would like to use December export figures from a particular country to predict total exports from that country in the coming year.

a) Use the data to come up with an equation for predicting total 2001 exports from December 2000 exports.

b) Assume that the equation you found in part "a" will apply to any year. If the coffee export of a country in December is 100, within what range would we expect to find the total exports for that country to fall in the coming year.

next page

Data Set 2 (Download Excel File "Salaries")

Data set 2 has gender, job class (1 through 8), and salary of 382 employees of a certain business. Below is a plot of average salaries for the males and females for the various job classifications. The plot suggest two things: (1) there appears to be a curved relationship between job class and salary, and (2) there appears to be differences between average salaries of males and females particularly at the upper job class levels.

a) Fit a multiple regression model to the data to predict salary Y from the gender indicator variable X1, (1 = male, 0 = female), job class X2, and the square of the job class X3 = (X2)2. You will have to create X1 and X3 yourself.

b) Based on your analysis, do you think gender has a significant effect on salary? What in the output tells you this?

c) Based on you analysis, do you think that the quadratic term ought to be included in the equation? What in the analysis tells you this?

next page

Data Set 3 (Download Excel File "Demand Deposits")

Data set 3 has demand deposits at commercial banks from Jan 2002 through Mar 2004. The plot is shown below.

a. Compute the exponentially smoothed forecast and the forecast errors. Use the weight w = .5.

b. Plot the data and the forecast on the same plot.

c. Comment on how well the exponentially smoothed forecast actually forecasts the demand deposits. 

• Attachment 1
• Attachment 2
• Attachment 3
• Attachment 4