Research for Applied Statistics

Syndicate content
Polynomial Families of Distributions
March 1, 1992
It is often necessary to estimate the population distribution of a random variate from a sample of observed values. Standard parametric families may not provide satisfactory fit to the data. A polynomial family is constructed by assuming that the distribution function G is a constrained polynomial of the cumulative distribution F of a convenient parametric family. Polynomial families offer great flexibility in data fitting, while retaining the important feature of parametric families that information in the data is condensed into a moderate number of values. This research contribution presents some theory of polynomial families.
Read More | Download Report
Robust Estimation Using Student's t Distribution
December 1, 1990
Robust estimators are useful when observations contain gross errors or are sampled from a heavy-tailed distribution. Student's t distributions with small degrees of freedom have heavy tails. Therefore, maximum likelihood estimation using these distributions provides simultaneous robust estimates of location and scale. In addition, the likelihood values can be used to choose among the available t distributions, making it unnecessary to make a subjective choice of an estimator. Monte Carlo results show these estimators to be as efficient as the bi-weight estimators of location and scale.
Read More | Download Report
A Maximum Likelihood Procedure for Developing a Common Metric in Item-Response Theory
February 1, 1989
Because the ability scale in item-response theory is arbitrary, if two item pools are calibrated in two different samples, their parameter estimates must be placed on a common metric using items administered in both calibrations. In this memorandum, a maximum-likelihood procedure for doing so is illustrated.
Read More | Download Report
Are Satisfaction and Dissatisfaction Really Opposites? Ordered versus Unordered Models of Satisfaction with Military Life
January 1, 1989
This paper discusses the degree to which satisfaction and dissatisfaction with military life represent directionally opposite aspects of the same basic phenomenon. The analysis approach is to estimate an unordered model of survey responses regarding satisfaction with military life from Marine respondents to the 1985 DOD Member Survey. These estimates can be used to test whether satisfaction and dissatisfaction are opposites with respect to a particular set of variables.
Read More | Download Report
Statistical Methods for Improving the Assessment of Recruiting Market Conditions
June 1, 1988
This research memorandum quantifies historical seasonal patterns in recruiting statistics and shows them to have evolved over time. Decline in seasonal variation dates from 1978 to 1980 for various measures of recruiting flows. Deseasonalized versions of these recruiting statistics are produced for use in econometric time-series models. The results provide a benchmark for gauging the seasonal component of Navy recruiting goals and achievements.
Read More | Download Report
Retention of Navy Veterans in the Selected Reserve
May 1, 1988
This analysis develops and estimates a model of retention during the first year of affiliation in the Selected Reserve. Estimates of the effects of pay and personal characteristics are provided for 11 Navy rating groups. The results should be useful for forecasting Selected Reserve manpower levels and for evaluating the effects of affiliation and retention bonuses.
Read More | Download Report
Assessment of DON Shore Base Readiness Analysis
May 1, 1988
This research memorandum documents the Center for Naval Analyses' assessment of the Price Waterhouse shore base facility condition readiness model. The accuracy and reasonableness of the model's predictions are assessed. Suggestions are made for revising the presentation of the model's results and for refining the model.
Read More | Download Report
Learning Curves, Personal Characteristics, and Job Performance
March 1, 1988
This research contribution uses data on the productivity of Naval Reserve recruiters to estimate the effects of on-the-job learning, experience, and individual characteristics on job performance. The econometric approach begins with the Poisson distribution, whose mean is assumed to be a function of explanatory variables. Generalizations are specified to control for individual heterogeneity as well as over-dispersion.
Read More | Download Report
Relations Between Sustaining Engineering, Methods of Management, and System Quality
December 1, 1987
The relations between Sustaining Engineering (SE) usage, methods of management, and system quality are determined in this research memorandum. Principal components analysis was used to combine multiple measures of effectiveness into a single measure of aircraft quality and multiple program characteristics into a single measure of high-level management. Levels of SE usage were used with these measures to determine the relations, if any.
Read More | Download Report
Simulating Cable Logging in the Alleghany Region of Pennsylvania: A Case Study
August 1, 1987
The eight-county Allegheny Region of Pennsylvania is 84% forested and comprises one of the state's most valuable timbersheds. Much of this area is inaccessible to conventional logging equipment due to steep slopes, poor drainage, and high erodability. For this reason, there has been a recently growing interest in using cable yarders to harvest these areas. A computer simulation model has been developed to aid in investigating the feasibility of cable logging in this region. This paper explains the design of the model, discusses the field data collected, evaluates the model, and makes recommendations for further data acquisition and model refinement.
Read More | Download Report