The Commander of Northwest Air Force Base (NAB) Is Struggling with a Rise of Driving While

PROJECT 5

The Commander of Northwest Air Force Base (NAB) is struggling with a rise of driving while intoxicated/driving under the influences (DUI) on her base and within her command. She asks her analyst to pull some historical data and provide her some information on an intervention strategy that was implemented by one of her predecessors. The analysts went through the records held by the Law Enforcement Command and found that in January of 2005, General John E. Walker implemented a “Commander’s Against Drunk Driving (CADD)” program in which he held all subordinate commanders at the Flight level responsible for the DUI within their respective units. Analyst looked at records from the three years prior to the Gen. Walkers intervention and the three years after the intervention (see attached Excel File) and constructed a time series analysis of the data. The results of the time series run is listed below.

Use the Check list (Table 18.26) found on page 18-61 of the text book (.pdf addition) and the information provided below to write a report analyzing the time series data provided in the attached Excel file explaining to the commander of NAB your findings and what the ramifications are if she took similar actions. (Hint: Report should be similar in scope to “Results at the end of Chapter 18”. You may also want to run some excel programs to further examine the data.

Case Processing Summary
No. of DUI Arrest
Series or Sequence Length / 72
Number of Missing Values in the Plot / User-Missing / 0
System-Missing / 0

Case Processing Summary
No. of DUI Arrest
Series or Sequence Length / 72
Number of Missing Values in the Plot / User-Missing / 0
System-Missing / 0

Model Description
Model Type
Model ID / No. of DUI Arrest / Model_1 / ARIMA(1,0,12)

Model Summary

Model Fit
Fit Statistic / Mean / SE / Minimum / Maximum / Percentile
5 / 10 / 25 / 50 / 75 / 90 / 95
Stationary R-squared / .563 / . / .563 / .563 / .563 / .563 / .563 / .563 / .563 / .563 / .563
R-squared / .563 / . / .563 / .563 / .563 / .563 / .563 / .563 / .563 / .563 / .563
RMSE / 2.758 / . / 2.758 / 2.758 / 2.758 / 2.758 / 2.758 / 2.758 / 2.758 / 2.758 / 2.758
MAPE / 46.128 / . / 46.128 / 46.128 / 46.128 / 46.128 / 46.128 / 46.128 / 46.128 / 46.128 / 46.128
MaxAPE / 463.340 / . / 463.340 / 463.340 / 463.340 / 463.340 / 463.340 / 463.340 / 463.340 / 463.340 / 463.340
MAE / 2.007 / . / 2.007 / 2.007 / 2.007 / 2.007 / 2.007 / 2.007 / 2.007 / 2.007 / 2.007
MaxAE / 8.257 / . / 8.257 / 8.257 / 8.257 / 8.257 / 8.257 / 8.257 / 8.257 / 8.257 / 8.257
Normalized BIC / 2.267 / . / 2.267 / 2.267 / 2.267 / 2.267 / 2.267 / 2.267 / 2.267 / 2.267 / 2.267
Model Statistics
Model / Number of Predictors / Model Fit statistics / Ljung-Box Q(18) / Number of Outliers
Stationary R-squared / R-squared / Statistics / DF / Sig.
No. of DUI Arrest-Model_1 / 0 / .563 / .563 / 20.966 / 15 / .138 / 0

Model Description
Model Name / MOD_2
Series or Sequence / 1 / No. of DUI Arrest
Transformation / None
Non-Seasonal Differencing / 1
Seasonal Differencing / 0
Length of Seasonal Period / No periodicity
Horizontal Axis Labels / Actual Month
Intervention Onsets / None
Reference Lines / None
Area Below the Curve / Not filled
Applying the model specifications from MOD_2
Case Processing Summary
No. of DUI Arrest
Series or Sequence Length / 72
Number of Missing Values in the Plot / User-Missing / 0
System-Missing / 0

* Sequence Charts.

TSPLOT VARIABLES=DUIARREST

/ID=ActualMonth

/LN

/DIFF=1

/FORMAT NOFILL NOREFERENCE.

Case Processing Summary
No. of DUI Arrest
Series or Sequence Length / 72
Number of Missing Values in the Plot / Negative or Zero Before Log Transform / 1a
User-Missing / 0
System-Missing / 0
a. The minimum value is .000.

ACF VARIABLES=DUIARREST

/LN

/DIFF=1

/MXAUTO 16

/SERROR=IND

/PACF.

Case Processing Summary
No. of DUI Arrest
Series Length / 72
Number of Missing Values / Negative or Zero Before Log Transform / 1a
User-Missing / 0
System-Missing / 0
Number of Valid Values / 71
Number of Values Lost Due to Differencing / 1
Number of Computable First Lags After Differencing / 67
a. The minimum value is .000
Autocorrelations
Series: No. of DUI Arrest
Lag / Autocorrelation / Std. Errora / Box-Ljung Statistic
Value / df / Sig.b
1 / -.054 / .117 / .216 / 1 / .642
2 / -.259 / .115 / 5.253 / 2 / .072
3 / -.473 / .114 / 22.388 / 3 / .000
4 / .144 / .113 / 24.004 / 4 / .000
5 / .074 / .112 / 24.433 / 5 / .000
6 / .185 / .112 / 27.185 / 6 / .000
7 / .096 / .111 / 27.936 / 7 / .000
8 / .088 / .110 / 28.575 / 8 / .000
9 / -.320 / .110 / 37.090 / 9 / .000
10 / -.332 / .110 / 46.258 / 10 / .000
11 / .132 / .109 / 47.720 / 11 / .000
12 / .462 / .108 / 66.079 / 12 / .000
13 / .201 / .107 / 69.608 / 13 / .000
14 / -.347 / .106 / 80.316 / 14 / .000
15 / -.196 / .105 / 83.804 / 15 / .000
16 / -.090 / .104 / 84.556 / 16 / .000
a. The underlying process assumed is independence (white noise).
b. Based on the asymptotic chi-square approximation.
Partial Autocorrelations
Series: No. of DUI Arrest
Lag / Partial Autocorrelation / Std. Error
1 / -.054 / .120
2 / -.262 / .120
3 / -.544 / .120
4 / -.096 / .120
5 / -.301 / .120
6 / -.165 / .120
7 / .154 / .120
8 / .237 / .120
9 / -.055 / .120
10 / -.296 / .120
11 / -.055 / .120
12 / .113 / .120
13 / .169 / .120
14 / -.065 / .120
15 / .125 / .120
16 / -.050 / .120

Regression

Descriptive Statistics
Mean / Std. Deviation / N
No. of DUI Arrest / 6.18 / 4.081 / 72
Month / 36.50 / 20.928 / 72
Correlations
No. of DUI Arrest / Month
Pearson Correlation / No. of DUI Arrest / 1.000 / -.646
Month / -.646 / 1.000
Sig. (1-tailed) / No. of DUI Arrest / . / .000
Month / .000 / .
N / No. of DUI Arrest / 72 / 72
Month / 72 / 72
Variables Entered/Removeda
Model / Variables Entered / Variables Removed / Method
1 / Monthb / . / Enter
a. Dependent Variable: No. of DUI Arrest
b. All requested variables entered.
Model Summaryb
Model / R / R Square / Adjusted R Square / Std. Error of the Estimate
1 / .646a / .418 / .410 / 3.136
a. Predictors: (Constant), Month
b. Dependent Variable: No. of DUI Arrest
Coefficientsa
Model / Unstandardized Coefficients / Standardized Coefficients / t / Sig.
B / Std. Error / Beta
1 / (Constant) / 10.782 / .747 / 14.435 / .000
Month / -.126 / .018 / -.646 / -7.089 / .000
a. Dependent Variable: No. of DUI Arrest
Coefficient Correlationsa
Model / Month
1 / Correlations / Month / 1.000
Covariances / Month / .000
a. Dependent Variable: No. of DUI Arrest
Residuals Statisticsa
Minimum / Maximum / Mean / Std. Deviation / N
Predicted Value / 1.71 / 10.66 / 6.18 / 2.638 / 72
Residual / -6.404 / 7.613 / .000 / 3.114 / 72
Std. Predicted Value / -1.696 / 1.696 / .000 / 1.000 / 72
Std. Residual / -2.042 / 2.428 / .000 / .993 / 72
a. Dependent Variable: No. of DUI Arrest

Charts

Regression (Baseline Months Jan. 2002 – Dec. 2004)

Descriptive Statistics
Mean / Std. Deviation / N
No. of DUI Baseline / 8.86 / 3.914 / 36
Baseline Months / 18.50 / 10.536 / 36
Correlations
No. of DUI Baseline / Baseline Months
Pearson Correlation / No. of DUI Baseline / 1.000 / -.116
Baseline Months / -.116 / 1.000
Sig. (1-tailed) / No. of DUI Baseline / . / .250
Baseline Months / .250 / .
N / No. of DUI Baseline / 36 / 36
Baseline Months / 36 / 36
Variables Entered/Removeda
Model / Variables Entered / Variables Removed / Method
1 / Baseline Monthsb / . / Enter
a. Dependent Variable: No. of DUI Baseline
b. All requested variables entered.
Model Summaryb
Model / R / R Square / Adjusted R Square / Std. Error of the Estimate
1 / .116a / .013 / -.016 / 3.945
a. Predictors: (Constant), Baseline Months
b. Dependent Variable: No. of DUI Baseline
ANOVAa
Model / Sum of Squares / df / Mean Square / F / Sig.
1 / Regression / 7.222 / 1 / 7.222 / .464 / .500b
Residual / 529.084 / 34 / 15.561
Total / 536.306 / 35
a. Dependent Variable: No. of DUI Baseline
b. Predictors: (Constant), Baseline Months
Coefficientsa
Model / Unstandardized Coefficients / Standardized Coefficients / t / Sig.
B / Std. Error / Beta
1 / (Constant) / 9.659 / 1.343 / 7.193 / .000
Baseline Months / -.043 / .063 / -.116 / -.681 / .500
a. Dependent Variable: No. of DUI Baseline
Coefficient Correlationsa
Model / Baseline Months
1 / Correlations / Baseline Months / 1.000
Covariances / Baseline Months / .004
a. Dependent Variable: No. of DUI Baseline
Residuals Statisticsa
Minimum / Maximum / Mean / Std. Deviation / N
Predicted Value / 8.11 / 9.62 / 8.86 / .454 / 36
Residual / -6.055 / 7.160 / .000 / 3.888 / 36
Std. Predicted Value / -1.661 / 1.661 / .000 / 1.000 / 36
Std. Residual / -1.535 / 1.815 / .000 / .986 / 36
a. Dependent Variable: No. of DUI Baseline

Charts

Regression (Post Intervention Months Jan. 2005 – Dec. 2007)

Descriptive Statistics
Mean / Std. Deviation / N
No. of DUI Baseline / 8.86 / 3.914 / 36
Baseline Months / 18.50 / 10.536 / 36
Correlations
No. of DUI Baseline / Baseline Months
Pearson Correlation / No. of DUI Baseline / 1.000 / -.116
Baseline Months / -.116 / 1.000
Sig. (1-tailed) / No. of DUI Baseline / . / .250
Baseline Months / .250 / .
N / No. of DUI Baseline / 36 / 36
Baseline Months / 36 / 36
Variables Entered/Removeda
Model / Variables Entered / Variables Removed / Method
1 / Baseline Monthsb / . / Enter
a. Dependent Variable: No. of DUI Baseline
b. All requested variables entered.
Model Summaryb
Model / R / R Square / Adjusted R Square / Std. Error of the Estimate
1 / .116a / .013 / -.016 / 3.945
a. Predictors: (Constant), Baseline Months
b. Dependent Variable: No. of DUI Baseline
ANOVAa
Model / Sum of Squares / df / Mean Square / F / Sig.
1 / Regression / 7.222 / 1 / 7.222 / .464 / .500b
Residual / 529.084 / 34 / 15.561
Total / 536.306 / 35
a. Dependent Variable: No. of DUI Baseline
b. Predictors: (Constant), Baseline Months
Coefficientsa
Model / Unstandardized Coefficients / Standardized Coefficients / t / Sig.
B / Std. Error / Beta
1 / (Constant) / 9.659 / 1.343 / 7.193 / .000
Baseline Months / -.043 / .063 / -.116 / -.681 / .500
a. Dependent Variable: No. of DUI Baseline
Coefficient Correlationsa
Model / Baseline Months
1 / Correlations / Baseline Months / 1.000
Covariances / Baseline Months / .004
a. Dependent Variable: No. of DUI Baseline
Residuals Statisticsa
Minimum / Maximum / Mean / Std. Deviation / N
Predicted Value / 8.11 / 9.62 / 8.86 / .454 / 36
Residual / -6.055 / 7.160 / .000 / 3.888 / 36
Std. Predicted Value / -1.661 / 1.661 / .000 / 1.000 / 36
Std. Residual / -1.535 / 1.815 / .000 / .986 / 36
a. Dependent Variable: No. of DUI Baseline

Charts