Your manager has asked you to predict repair parts purchases for next year given
ID: 455795 • Letter: Y
Question
Your manager has asked you to predict repair parts purchases for next year given estimates of the following values: time period in years, sales, current maintenance cycle in months, average machine age, and predominant machine manufacturer (there are four different brands in the full data set). An excerpt of the 15 years of data is shown below. Correlations between Repair Parts Purchases and the quantitative variables are shown on the last row. Based on this information, choose the approach that is likely to be the most effective from the selection below.
Based on this information, it appears that the best variables are maintenance cycle and average machine age. So, I will build a regression model that uses these two variables to predict Repair Parts Purchases for next year.
Other than time period (year), none of the other causal variables look useful in this case. So, I will build a trend projection model using year as the independent variable to predict Repair Parts Purchases for next year.
Based on this information, it appears that the best variables are year and sales. So, I will build a regression model that uses these two variables to predict Repair Parts Purchases for next year.
I will create a set of indicator (dummy) variables for Predominant Machine Brand. Then, I will build a regression model using all but one of the indicator variables and the other quanitative variables to predict Repair Parts Purchases. In an interative fashion, I will eliminate the variable with the highest p-value and continue running regression on the remaining variables until the model contains only significant variables.
Other than time period (year), the other quantitative variables do not look useful in this case. But Predominant Machine Brand may be helpful. So, I will create a set of indicator (dummy) variables for Predominant Machine Brand. Then, I will build a regression model using all but one of the indicator variables and time. In an interative fashion, I will eliminate the independent variable with the highest p-value and continue running regression on the remaining variables until the model contains only significant variables.
a.Based on this information, it appears that the best variables are maintenance cycle and average machine age. So, I will build a regression model that uses these two variables to predict Repair Parts Purchases for next year.
b.Other than time period (year), none of the other causal variables look useful in this case. So, I will build a trend projection model using year as the independent variable to predict Repair Parts Purchases for next year.
c.Based on this information, it appears that the best variables are year and sales. So, I will build a regression model that uses these two variables to predict Repair Parts Purchases for next year.
d.I will create a set of indicator (dummy) variables for Predominant Machine Brand. Then, I will build a regression model using all but one of the indicator variables and the other quanitative variables to predict Repair Parts Purchases. In an interative fashion, I will eliminate the variable with the highest p-value and continue running regression on the remaining variables until the model contains only significant variables.
e.Other than time period (year), the other quantitative variables do not look useful in this case. But Predominant Machine Brand may be helpful. So, I will create a set of indicator (dummy) variables for Predominant Machine Brand. Then, I will build a regression model using all but one of the indicator variables and time. In an interative fashion, I will eliminate the independent variable with the highest p-value and continue running regression on the remaining variables until the model contains only significant variables.
Average Predominant Maintenance Cycle months 6 6 6 6 6 12 Machine Age 3.5 4.5 5.2 6 2.1 2.8 Machine Brand CM CM CM CM AP AP Repair Part Purchases Sales 733 715 631 730 795 726 Year 102610 115060 119780 137030 137650 146090 4 6 Correlation with Purchases 0.03467318 0.465936233 0.04532786 0.33535958Explanation / Answer
Based on this information, it appears that the best variables are maintenance cycle and average machine age becase they have the largest correlation coefficient among all others
So, I will build a regression model that uses these two variables to predict Repair Parts Purchases for next year.
Hence, the correct option is option-A
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.