Regression models of quarterly indirect labor hours for NARF Alameda/

• Main Author: Woodward, Michael R.
• Other Authors: NA
• Language: en_US
• Published: 2012
• Online Access: http://hdl.handle.net/10945/22453

Since overhead costs for indirect labor account for a large percentage of the Naval Air Rework Facility's (NARF) total budget, it is essential that management be able to predict these costs accurately. The research performed in this thesis uses data from the major cost centers which comprise NARF Alameda. Regression models of their indirect labor to be used for forecasting purposes were developed. Quarterly data were used in the analysis, requiring transformation of the data to eliminate the effects of autocorrelation. The Durbin-Watson test was used to check for the effects of first-order autocorrelation and Wallis' test was used for fourth-order autocorrelation. Once the effects of autocorrelation were eliminated, excellent structural results were obtained for twelve of the thirteen cost centers of interest. Predictive analysis performed using withheld data showed the final models can be expected to yield reliable forecasts