UnitsMinutes1232293494644745876966977109811991499145101541016611162111741218012176141791619317193181951819820205 CCM DATA Columbia Computer Maintenance (CCM) is a venture to establish a new service organization for the on-site maintenance of personal computers and related peripheral devices. The management of CCM isconcerned with developing pricingschedules and planning future staffing levels for customer serviceengineers.Staffing level requirement forecasts will be constructed based upon two principal components - thedemand for service as measured by the number of service callsand the length of a typical service call.Establishing baselines for the latter component, the length of service calls, is the subject of this analysis.The length of a service call appears to depend, quite naturally, upon the number ofunits/devices(computers and/or peripherals) to be repaired or replaced during thevisit. In order to establish thenature of the relationship between length of call andnumber of units to be serviced, a random sampleof n = 24 service call records hasbeen collected for analysis. The data comprise the length of the servicecall in minutes and the number of units/devices serviced.Your task is to perform a simple linear regression analysis of the service call data and to interpret theresults by addressing the issued posed in exercise parts outlined below. An important objective of thisexercise is to give you the opportunity to familiarize yourself with statistical modeling computingresources available through Microsoft Excel and JASP - which, including R will be useful computationalresources for your final modeling projects. The data for this exerciseare included as attachments to thisassignment and are cross-posted to the Exercises folder of our WISE course site-WISE GSM 5103 Resources Exercises - CCMData.xlsx for use with Excel, CCMDato.csv, for use with JASP andR. A complete listing of the data follows.Units12344566Minutes2329496474879697Units79910101111Minutes109119149145154166162174Units1212141617181820ExcelComplete the following exercise parts using Excel and the Excel version of the data,CCMData.xlsx.Minutes180176179193193195198205a. Data Understanding: Compute a complete set of descriptive statistics for each of the twovariables - units and minutes. Examine the relationship between length of service time,minutes, and number of components repaired, units, by constructing an appropriate scatterplot of the data. Howis minutes apparently related to units? Compute the correlationcoefficient between minutes and units [CORREL(ynite, minutes)). Do you expect that units willbe a relatively good predictor of minutes? b. Estimation: Fit the simple linear regression model for the response, minutes, on thepredictor, units; that is, estimate the coefficients of the simple linear regression modelexpressing minutes as a linear function of units. Try computing the estimated regressioncoefficients using each of the following three approaches.Trend line added to a scatter plot INTERCEPT andSLOPE built-in functionsThe Regression procedure in the Excel Data Analysis Tool Pack.Interpret the estimated parameters to make a statement, in managerial terms, about theapparent fixed and variable components of the service time.C. Summary Measures: Consider the analysis of variance table from your JASP analysis.Interpret the F-test for model significance. Conduct your test at the a = 0.01 level ofsignificance. State your conclusion in managerial terms, reporting a p-value for your test.Identify and interpret the coefficientof determination, R², and the residual standard deviation,s. Based upon the results of the F-test and the model R2, would you conclude that the modelprovides a reasonable fit to the data?d. Parameter Estimates / Model Coefficients: Interpret significance tests for the individualmodel coefficients 30 and 31. Conduct your tests at thea = 0.01 level of significance. Report thep-values for the tests. Compute 95% confidence intervals for the model coefficients 30 and 31.Interpret the estimated coefficients to contribute managerial insight into the apparent fixedand variable components of the service time.e. Diagnostic Checking: Construct a plot of the model residuals against the fitted values and anormal probability plot (Q-Q plot) of the (standardized) residuals. Use these plots to assess theadequacy of the fittedmodel; that is, to determine whether or not the required residualassumptions hold. Does the simple linear regression model appear to be adequate (does itsatisfy the four residual assumptions)? Explain. In light ofyour conclusions regarding modeladequacy, would you offer any warnings about your predictions of part e? Can you offer anysuggestions as to how the current simple linear regression model might be improved?f. Predictions: Use your model to predict the number of minutes required to repair x units forthe x values 10, 20, and 30, respectively. Although neither Excel nor JASP offer built-inprediction capability, point predictions can be easily produced using Excel as a scratch pad (or,for that matter, with nothing more than paper and pencil,) but proper interval estimates can bea bit more challenging. For now, it will suffice to compute crude approximate predictionintervals based on the residual standard error, s.What to submit: Prepare a brief summary document (target size, two pages) ofyour findings, in the formof a Microsoft Word document, addressing directly the questions posed in each problem part. Alsosubmit a copy of an Excel workbook documenting your work for parts (a) and (b), and a copy of the JASPanalysis file produced to support your analysis of parts (c)- (f).

Question

UnitsMinutes1232293494644745876966977109811991499145101541016611162111741218012176141791619317193181951819820205 CCM DATA Columbia Computer Maintenance (CCM) is a venture to establish a new service organization for the on-site maintenance of personal computers and related peripheral devices. The management of CCM isconcerned with developing pricingschedules and planning future staffing levels for customer serviceengineers.Staffing level requirement forecasts will be constructed based upon two principal components - thedemand for service as measured by the number of service callsand the length of a typical service call.Establishing baselines for the latter component, the length of service calls, is the subject of this analysis.The length of a service call appears to depend, quite naturally, upon the number ofunits/devices(computers and/or peripherals) to be repaired or replaced during thevisit. In order to establish thenature of the relationship between length of call andnumber of units to be serviced, a random sampleof n = 24 service call records hasbeen collected for analysis. The data comprise the length of the servicecall in minutes and the number of units/devices serviced.Your task is to perform a simple linear regression analysis of the service call data and to interpret theresults by addressing the issued posed in exercise parts outlined below. An important objective of thisexercise is to give you the opportunity to familiarize yourself with statistical modeling computingresources available through Microsoft Excel and JASP - which, including R will be useful computationalresources for your final modeling projects. The data for this exerciseare included as attachments to thisassignment and are cross-posted to the Exercises folder of our WISE course site-WISE > GSM 5103 >Resources > Exercises - CCMData.xlsx for use with Excel, CCMDato.csv, for use with JASP andR. A complete listing of the data follows.Units12344566Minutes2329496474879697Units79910101111Minutes109119149145154166162174Units1212141617181820ExcelComplete the following exercise parts using Excel and the Excel version of the data,CCMData.xlsx.Minutes180176179193193195198205a. Data Understanding: Compute a complete set of descriptive statistics for each of the twovariables - units and minutes. Examine the relationship between length of service time,minutes, and number of components repaired, units, by constructing an appropriate scatterplot of the data. Howis minutes apparently related to units? Compute the correlationcoefficient between minutes and units [CORREL(ynite, minutes)). Do you expect that units willbe a relatively good predictor of minutes? b. Estimation: Fit the simple linear regression model for the response, minutes, on thepredictor, units; that is, estimate the coefficients of the simple linear regression modelexpressing minutes as a linear function of units. Try computing the estimated regressioncoefficients using each of the following three approaches.Trend line added to a scatter plot INTERCEPT andSLOPE built-in functionsThe Regression procedure in the Excel Data Analysis Tool Pack.Interpret the estimated parameters to make a statement, in managerial terms, about theapparent fixed and variable components of the service time.C. Summary Measures: Consider the analysis of variance table from your JASP analysis.Interpret the F-test for model significance. Conduct your test at the a = 0.01 level ofsignificance. State your conclusion in managerial terms, reporting a p-value for your test.Identify and interpret the coefficientof determination, R², and the residual standard deviation,s. Based upon the results of the F-test and the model R2, would you conclude that the modelprovides a reasonable fit to the data?d. Parameter Estimates / Model Coefficients: Interpret significance tests for the individualmodel coefficients 30 and 31. Conduct your tests at thea = 0.01 level of significance. Report thep-values for the tests. Compute 95% confidence intervals for the model coefficients 30 and 31.Interpret the estimated coefficients to contribute managerial insight into the apparent fixedand variable components of the service time.e. Diagnostic Checking: Construct a plot of the model residuals against the fitted values and anormal probability plot (Q-Q plot) of the (standardized) residuals. Use these plots to assess theadequacy of the fittedmodel; that is, to determine whether or not the required residualassumptions hold. Does the simple linear regression model appear to be adequate (does itsatisfy the four residual assumptions)? Explain. In light ofyour conclusions regarding modeladequacy, would you offer any warnings about your predictions of part e? Can you offer anysuggestions as to how the current simple linear regression model might be improved?f. Predictions: Use your model to predict the number of minutes required to repair x units forthe x values 10, 20, and 30, respectively. Although neither Excel nor JASP offer built-inprediction capability, point predictions can be easily produced using Excel as a scratch pad (or,for that matter, with nothing more than paper and pencil,) but proper interval estimates can bea bit more challenging. For now, it will suffice to compute crude approximate predictionintervals based on the residual standard error, s.What to submit: Prepare a brief summary document (target size, two pages) ofyour findings, in the formof a Microsoft Word document, addressing directly the questions posed in each problem part. Alsosubmit a copy of an Excel workbook documenting your work for parts (a) and (b), and a copy of the JASPanalysis file produced to support your analysis of parts (c)- (f).

Accepted Answer

Given Information:
Consider the given dataset:

Units
Minutes

1
23

2
29

3
49

4
64

4
74…

Units	Minutes
1	23
2	29
3	49
4	64
4	74
5	87
6	96
6	97
7	109
8	119
9	149
9	145
10	154
10	166
11	162
11	174
12	180
12	176
14	179
16	193
17	193
18	195
18	198
20	205