XYZ logistics is developing accost formula for its packing activity. Management has identified that packing cost may be driven by # of customer orders and/or the size of an order (weight in lbs.). You have been given the following data.

Assume your role as a management accountant who has recently learn regression analysis and is interested in further understanding the role of predictive analytics in cost management.

Required:

(1). Evaluate the possible cost functions using the following criteria, (a) economic plausibility, (b) Goodness-of-fit and (c) individual tests (t tests) for the cost parameters at 95% level of accuracy. Use the following format for the analysis.

Criterion	Model (1)	.	Model (N)
a) Economic plausibility
b) Goodness-of-fit
c) Individual t tests for cost parameters

(2). Which model would you recommend and why?

(3). For the model with two drivers, predict the packing cost for 25000 orders weighing 40,000 lbs.

(4). Prepare a 99% confidence interval for the forecast packing cost in part (3) above.

Part B: (5 marks)

Clearly explain with details the confidence interval obtained in (part 4)?

\begin{tabular}{|r|r|r|r|} \hline Month & Packing Cost in $ & Number of Orders & Weight of Orders \\ \hline 1 & 45000 & 11200 & 24640 \\ \hline 2 & 58000 & 14000 & 31220 \\ \hline 3 & 39000 & 10500 & 18000 \\ \hline 4 & 35600 & 9000 & 19350 \\ \hline 5 & 90000 & 21000 & 46200 \\ \hline 6 & 126000 & 31000 & 64000 \\ \hline 7 & 90600 & 20000 & 60000 \\ \hline 8 & 63000 & 15000 & 40000 \\ \hline 9 & 79000 & 16000 & 59000 \\ \hline 10 & 155000 & 40000 & 88000 \\ \hline 11 & 450000 & 113500 & 249700 \\ \hline 12 & 640000 & 150000 & 390000 \\ \hline 13 & 41000 & 10000 & 23000 \\ \hline 14 & 54000 & 14000 & 29400 \\ \hline 15 & 58000 & 15000 & 30000 \\ \hline 16 & 58090 & 14500 & 31900 \\ \hline 17 & 80110 & 18000 & 50000 \\ \hline 18 & 123000 & 30000 & 75000 \\ \hline 19 & 108000 & 27000 & 63450 \\ \hline 20 & 76000 & 18000 & 41400 \\ \hline \end{tabular}