Excel file attached

1 answer below »
Answered 1 days AfterMay 13, 2021

Answer To: Excel file attached

Saravana answered on May 15 2021
146 Votes
Ques 7.13.docx
Ques 7.13
A .Estimate the regression equation with months since last service as the only independent variable. Express the relationship in an equation. Test the coefficient at 0.05 level of significance. Interpret it. What is the coefficient of determination? Interpret it.
        SUMMARY OUTPUT
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        Regression Statistics
        
        
        
        
        
        
        
        Multiple R
        0.730874
        
        
        
        
        
        
        
        R Square
        0.534177
        
        
        
        
        
        
        
        Adjusted R Square
        0.475949
        
        
        
        
        
        
        
        Standard Error
        0.781022
        
        
        
        
        
        
        
        Observations
        10
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        ANOVA
        
        
        
        
        
        
        
        
         
        df
        SS
        MS
        F
        Significance F
        
        
        
        Regression
        1
        5.596033
        5.596033
        9.173887
        0.016338
        
        
        
        Residual
        8
        4.879967
        0.609996
        
        
        
        
        
        Total
        9
        10.476
         
         
         
        
        
        
        
        
        
        
        
        
        
        
        
         
        Coefficients
        Standard Error
        t Stat
        P-value
        Lower 95%
        Upper 95%
        Lower 95.0%
        Upper 95.0%
        Intercept
        2.147273
        0.604977
        3.549344
        0.007517
        0.752193
        3.542353
        0.752193
        3.542353
        Months Since Last Service
        0.304132
        0.100412
        3.028842
        0.016338
        0.072582
        0.535683
        0.072582
        0.535683
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
Repair time (hours) = 2.14 + (Months since last service) * 0.30
Intercept: The interc
ept coefficient is 2.14 and the coefficient was significantly different from zero ( t(8) = 3.54, p = 0.0075). The repair time was around 2.14 irrespective of the months of last service.
The months since last service coefficient was 0.30 and this coefficient was significant (t(8) = 3.02, p = 0.016) .An increase of one month since last service increased the repair time by 0.30 hours.
The coefficient of determination measured as R squared = 0.53 mean that 53% of variance in repair time being explained by months since last service.
B. Use model in a to predict repair time in hours for the below data. Compute the residuals. Do you see any pattern in residuals related to type of repair or the repairperson? Create two scatter diagrams of actual repair time against months since last service.
        Repair Time in Hours
        Months Since Last Service
        Type of Repair
        Repairperson
        Predicted Repair Time in Hours
        Residuals
        1.8
        3
        Mechanical
        Donna Newton
        3.05
        -1.25
        3
        6
        Mechanical
        Donna Newton
        3.97
        -0.97
        4.2
        9
        Mechanical
        Bob Jones
        4.88
        -0.68
        2.9
        2
        Electrical
        Donna Newton
        2.75
        0.14
        2.9
        2
        Electrical
        Donna Newton
        2.75
        0.14
        4.8
        8
        Electrical
        Bob Jones
        4.58
        0.21
        4.8
        8
        Mechanical
        Bob Jones
        4.58
        0.21
        4.5
        6
        Electrical
        Donna Newton
        3.97
        0.52
        4.9
        7
        Electrical
        Bob Jones
        4.27
        0.62
        4.4
        4
        Electrical
        Bob Jones
        3.36
        1.03
The predicted times for mechanical repairs are consistently higher, so we see a pattern of negative residuals for mechanical repairs.
Scatter plot of Repair time vs Months since last service split across Mechanical and electrical type of repairs
Scatter plot of Repair time vs Months since last service split across different repair person Donna Newton and Bob Jones
C. Create a dummy variable for type of repair. Estimate a regression model with months since last service and type of repair as independent variables. Express the relationship in an equation. Test the parameters of the model at 0.05 level of confidence. Interpret the parameters. Find the coefficient of determination. Interpret it.
Dummy Coded Data:
        Repair Time in Hours
        Months Since Last Service
        Type of Repair Dummy Variable
        Type of Repair
        2.9
        2
        0
        Electrical
        3.0
        6
        1
        Mechanical
        4.8
        8
        0
        Electrical
        1.8
        3
        1
        Mechanical
        2.9
        2
        0
        Electrical
        4.9
        7
        0
        Electrical
        4.2
        9
        1
        Mechanical
        4.8
        8
        1
        Mechanical
        4.4
        4
        0
        Electrical
        4.5
        6
        0
        Electrical
Linear Regression output:
        SUMMARY OUTPUT
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        Regression Statistics
        
        
        
        
        
        
        
        Multiple R
        0.926928
        
        
        
        
        
        
        
        R Square
        0.859195
        
        
        
        
        
        
        
        Adjusted R Square
        0.818964
        
        
        
        
        
        
        
        Standard Error
        0.459048
        
        
        
        
        
        
        
        Observations
        10
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        ANOVA
        
        
        
        
        
        
        
        
         
        df
        SS
        MS
        F
        Significance F
        
        
        
        Regression
        2
        9.000923
        4.500461
        21.357
        0.001048
        
        
        
        Residual
        7
        1.475077
        0.210725
        
        
        
        
        
        Total
        9
        10.476
         
         
         
        
        
        
        
        
        
        
        
        
        
        
        
         
        Coefficients
        Standard Error
        t Stat
        P-value
        Lower 95%
        Upper 95%
        Lower 95.0%
        Upper 95.0%
        Intercept
        2.193189
        0.355761
        6.164787
        0.000461
        1.351948
        3.034429
        1.351948
        3.034429
        Months Since Last Service
        0.387616
        0.062565
        6.195396
        0.000447
        0.239673
        0.535559
        0.239673
        0.535559
        Type of Repair Dummy Variable
        -1.26269
        0.314127
        -4.0197
        0.005062
        -2.00549
        -0.5199
        -2.00549
        -0.5199
The months since last service is still has a positive coefficient = 0.38, the coefficient is also significant t(7) = 6.19 = 0.0004. So, for every increase in one month in delayed service increased the time spent in repair by 0.38 hours.
To interpret the dummy coded Type of repair variable we need to interpret the coefficient as the mean difference between Electrical repairs (dummy coded -0) and Mechanical Repairs (dummy coded as 1). Since the mean difference is negative we can infer that electrical repairs take longer than mechanical repairs. The t-test in this scenario is tests whether the mean difference between the two groups is significant or not. We can observe that the mean difference between the electrical and mechanical repair is significant (t(7) = -4.0197, p = 0.0052).
The coefficient of determination measured as R Square = 0.859195, means that 85% of the variance in repair times is explained by two predictors: months since last service and the type of repair (electrical or mechanical).
D. Create a dummy variable for repairperson. Estimate a regression model with months since last service and repairperson as independent variables. Express the relationship in an equation. Test the parameters of the model at 0.05 level of confidence. Interpret the parameters. Find the coefficient of determination. Interpret it.
The dummy coded data:
        Repair Time in Hours
        Months Since Last Service
        Repairperson Dummy Variable
        Repairperson
        2.9
        2
        1
        Donna Newton
        3
        6
        1
        Donna Newton
        4.8
        8
        0
        Bob Jones
        1.8
        3
        1
        Donna Newton
        2.9
        2
        1
        Donna Newton
        4.9
        7
        0
        Bob Jones
        4.2
        9
        0
        Bob Jones
        4.8
        8
        0
        Bob Jones
        4.4
        4
        0
        Bob Jones
        4.5
        6
        1
        Donna Newton
Repair person: Donna Newton is coded as 1 and Bob jones as 0
Linear regression output:
        SUMMARY OUTPUT
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        Regression Statistics
        
        
        
        
        
        
        
        Multiple R
        0.824936
        
        
        
        
        
        
        
        R Square
        0.680519
        
        
        
        
        
        
        
        Adjusted R Square
        0.589238
        
        
        
        
        
        
        
        Standard Error
        0.691467
        
        
        
        
        
        
        
        Observations
        10
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        ANOVA
        
        
        
        
        
        
        
        
         
        df
        SS
        MS
        F
        Significance F
        
        
        
        Regression
        2
        7.129114
        3.564557
        7.455258
        0.018431
        
        
        
        Residual
        7
        3.346886
        0.478127
        
        
        
        
        
        Total
        9
        10.476
         
         
         
        
        
        
        
        
        
        
        
        
        
        
        
         
        Coefficients
        Standard Error
        t Stat
        P-value
        Lower 95%
        Upper 95%
        Lower 95.0%
        Upper 95.0%
        Intercept
        3.526329
        0.93808
        3.75909
        0.007083
        1.308121
        5.744537
        1.308121
        5.744537
        Months Since Last Service
        0.151899
        0.123006
        1.234884
        0.256713
        -0.13897
        0.442763
        -0.13897
        0.442763
        Repairperson Dummy Variable
        -1.08354
        0.605112
        -1.79065
        0.116467
        -2.51441
        0.347318
        -2.51441
        0.347318
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
Though the whole regression model is significant (F(2,7) = 7.45, p = 0.018), We can find any significant individual predictors. The mean difference between Bob and Donna is negative indicating longer repair times for bob, but this mean difference was not significant (t(7) = 1.79, p = 0.1164).
The model has an R square = 0.68 meaning that the model as a whole explains 68% of variance in the repair times.
E. Estimate a regression model with months since last service, type of repai and repairperson as independent variables. Express the relationship in an equation. Test the parameters of the model at 0.05 level of confidence. Interpret the parameters. Find the coefficient of determination. Interpret it.
The dummy coded data:
        Repair Time in Hours
        Months Since Last Service
        Type of Repair Dummy Variable
        Repairperson Dummy Variable
        Type of Repair
        Repairperson
        2.9
        2
        0
        1
        Electrical
        Donna Newton
        3.0
        6
        1
        1
        Mechanical
        Donna Newton
        4.8
        8
        0
        0
        Electrical
        Bob Jones
        1.8
        3
        1
        1
        Mechanical
        Donna Newton
        2.9
        2
        0
        1
        Electrical
        Donna Newton
        4.9
        7
        0
        0
        Electrical
        Bob Jones
        4.2
        9
        1
        0
        Mechanical
        Bob Jones
        4.8
        8
        1
        0
        Mechanical
        Bob Jones
        4.4
        4
        0
        0
        Electrical
        Bob Jones
        4.5
        6
        0
        1
        Electrical
        Donna Newton
Linear regression Output:
        SUMMARY OUTPUT
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        Regression Statistics
        
        
        
        
        
        
        
        Multiple R
        0.948789
        
        
        
        
        
        
        
        R Square
        0.9002
        
        
        
        
        
        
        
        Adjusted R Square
        0.8503
        
        
        
        
        
        
        
        Standard Error
        0.417434
        
        
        
        
        
        
        
        Observations
        10
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        ANOVA
        
        
        
        
        
        
        
        
         
        df
        SS
        MS
        F
        Significance F
        
        
        
        Regression
        3
        9.430492
        3.143497
        18.04002
        0.002091
        
        
        
        Residual
        6
        1.045508
        0.174251
        
        
        
        
        
        Total
        9
        10.476
         
         
         
        
        
        
        
        
        
        
        
        
        
        
        
         
        Coefficients
        Standard Error
        t Stat
        P-value
        Lower 95%
        Upper 95%
        Lower 95.0%
        Upper 95.0%
        Intercept
        2.962567
        0.587176
        5.045452
        0.002344
        1.5258
        4.399334
        1.5258
        4.399334
        Months Since Last Service
        0.291444
        0.083598
        3.486238
        0.013043
        0.086886
        0.496002
        0.086886
        0.496002
        Type of Repair Dummy Variable
        -1.10241
        0.303344
        -3.63418
        0.010911
        -1.84466
        -0.36015
        -1.84466
        -0.36015
        Repairperson Dummy Variable
        -0.60909
        0.38793
        -1.5701
        0.167444
        -1.55832
        0.34014
        -1.55832
        0.34014
The months since last service is a significant predictor in the model (t(6) = 3.48, p = 0.013). The coefficient of Months since last service is 0.29. This indicates that a delay in 1 month increases the repair time by 0.29 hour.
The mean difference between the electrical and mechanical Type of repair is significant (t(6) = 3.63, p = 0.01). The mean difference between electrical and mechanical repair time is negative indicating larger repair times for electrical repairs compared to mechanical repairs.
There is no significant difference in the mean of the repair times of Bob and Donna (t(6) = -1.5701, p = 0.1674). The mean difference is negative, thus indicates a trend of longer repair time for Bob, but this trend is not significant.
The R squared of the model with three predictors is R-square = 0.9002. The model explains 90% of variance in repair times.
F. Which model would you use? Why?
The model with three predictors: Months since last service, Type of repair and Repairperson as an R square of 0.9002. Similarly, the two parameter model with Months since last service, Repairperson has in R square = 0.859 and finally the two parameter model with Months since last service, Type of repair has in R square = 0.68. And based on the largest R squared value we can choose the model with three predictors as best model.
Repair time (hours) vs Months since last service
Mechanical Repair    3    6    9    8    1.8    3    4.2    4.8    Electrical Repair    2    2    6    7    4    8    2.9    2.9    4.5    4.9000000000000004    4.4000000000000004    4.8    Months since last service
Repair time ()hours)
Repair Time (hours) vs Months since last service
Donna    3    6    2    2    6    1.8    3    2.9    2.9    4.5    Bob    9    8    8    7    4    4.2    4.8    4.8    4.9000000000000004    4.4000000000000004    Months since last service
Repair Time (hours)
Repair time (hours) vs Months since last service
Mechanical Repair    3    6    9    8    1.8    3    4.2    4.8    Electrical Repair    2    2    6    7    4    8    2.9    2.9    4.5    4.9000000000000004    4.4000000000000004    4.8    Months since last service
Repair time ()hours)
Repair Time (hours) vs Months since last service
Donna    3    6    2    2    6    1.8    3    2.9    2.9    4.5    Bob    9    8    8    7    4    4.2    4.8    4.8    4.9000000000000004    4.4000000000000004    Months since last service
Repair Time (hours)
Ques 7.18.docx
Ques 7.18:
A. Develop a scatter chart, treating asking rent as independent variable. Does a simple linear regression model appear to be appropriate?
B. Develop a simple linear regression to explain monthly mortgage. Express the relationship in an equation. Plot the residuals. Does a simple linear regression model appear to be appropriate based on the residual plot.
        SUMMARY OUTPUT
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        Regression Statistics
        
        
        
        
        
        
        
        Multiple R
        0.869565
        
        
        
        
        
        
        
        R Square
        0.756143
        
        
        
        
        
        
        
        Adjusted R Square
        0.725661
        
        
        
        
        
        
        
        Standard Error
        78.78191
        
        
        
        
        
        
        
        Observations
        10
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        
        ANOVA
        
        
        
        
        
        
        
        
         
        df
        SS
        MS
        F
        Significance F
        
        
        
        Regression
        1
        153961.7
        153961.7
        24.80616
        0.001079
        
        
        
        Residual
        8
        49652.72
        6206.59
        
        
        
        
        
        Total
        9
        203614.4
         
         
         
        
        
        
        
        
        
        
        
        
        
        
        
         
        Coefficients
        Standard Error
        t Stat
        P-value
        Lower 95%
        Upper 95%
        Lower 95.0%
        Upper 95.0%
        Intercept
        -197.958
        187.695
        -1.05468
        0.322379
        -630.784
        234.8671
        -630.784
        234.8671
        Rent ($)
        1.069929
        0.21482
        4.980579
        0.001079
        0.574553
        1.565305
        0.574553
        1.565305
Mortgage = -197.958 + (Rent) * 1.06
        RESIDUAL OUTPUT
        
        
        
        
        Observation
        Predicted Mortgage ($)
        Residuals
        1
        700.7819
        -161.782
        2
        938.306
        63.69396
        3
        682.5931
        -56.5931
        4
        635.5162
        75.4838
        5
        653.705
        1.295014
        6
        947.9354
        29.0646
        7
        821.6838
        -45.6838
        8
        712.5511
        -17.5511
        9
        617.3274
        33.67259
        10
        575.6002
        78.39981
Residual Plot:
The residual plot has a prominent feature:
1. There is a sort of curvilinear relation between the fitted vs residuals hinting at Non-linearity in data.
Thus, looking at the curvilinear relation in residual plot we can conclude that there is Non-linearity in the data will be best explained by a quadratic relationship rather than linear relationship in linear regression.
C. Develop a quadratic regression model. Express the relationship in an equation.
Quadratic regression...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here