Applied multivariate statistical analysis
Applied multivariate statistical analysis
Introduction to statistical pattern recognition (2nd ed.)
Introduction to statistical pattern recognition (2nd ed.)
An introduction to support Vector Machines: and other kernel-based learning methods
An introduction to support Vector Machines: and other kernel-based learning methods
Microsoft Excel 2000 Power Programming with VBA
Microsoft Excel 2000 Power Programming with VBA
R and S-Plus Companion to Applied Regression
R and S-Plus Companion to Applied Regression
Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond
Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond
Excel 2003 Power Programming with VBA
Excel 2003 Power Programming with VBA
A Comparison of Decision Tree Ensemble Creation Techniques
IEEE Transactions on Pattern Analysis and Machine Intelligence
Therapeutic Drug Monitoring of Kidney Transplant Recipients Using Profiled Support Vector Machines
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Improve Computer-Aided Diagnosis With Machine Learning Techniques Using Undiagnosed Samples
IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans
| Hi-index | 0.01 |
On April 10, 2006, Major General Bostick, the Commanding General of the United States Army Recruiting Command (USAREC), in conjunction with the Assistant Secretary of the Army for Manpower and Reserve Affairs, approved a new method of selecting individuals from within the Army for recruiting duty. Implementation currently waits for similar and current research concerning Drill Sergeants. Previously, the Army assigned successful noncommissioned officers (NCOs), regardless of inherent sales and marketing skills, into the recruiting force. The Army learned skills that make a successful combat leader do not always translate well into recruiting duty. This research began in 2001, after recruiting shortages, when the Army began researching recruiter selection methods. The result is an application that combines statistical learning with Industrial and Organizational (IO) psychology. The resulting selection model determines the better NCOs for service as detailed recruiters in the United States Army Recruiting Command, a 6000 plus sales force located worldwide. The application enhances IO psychology by providing a statistical prediction of job performance derived from psychological inventories and biographical data. The application uses a combination of statistical learning, variable selection methods, and IO psychology to determine the better prediction function approximation with variables obtained from the noncommissioned officer leadership skills inventory (NLSI) and biographical data. The application also creates a methodology for iteratively developing a statistical learning model. We learned that random forest models outperformed support vector regressions and stepwise regression for these data. A greedy algorithm enhanced model generalization by selecting a good subset of prediction variables. The model represents a multimodal relationship primarily between recruiter age, NLSI score, and, to a lesser degree, 34 other variables. The resulting model runs in R statistical language and is controlled within an Excel worksheet environment by using Visual Basic Application language and RExcel. The end product enables general utilization of a statistically elegant model, normally reserved for advanced researchers, engineers, statisticians, and economists.