| Estimating Percentiles from Environmental Samples when all Observations are Nondetectable |
JUN 96 |
12 pages |
| Authors:
Dennis E. Smith; Kevin C. Burns; DESMATICS INC STATE COLLEGE PA
|
 | Many environmental sampling problems involve some specified regulatory or contractual limit (RL). Often the interest is in estimating the percentile of the underlying contaminant concentration distribution corresponding to RL. In previous reports, we have discussed the problem of determining a lower 100(1-alpha)% confidence limit for that percentile when no observations are observable, but are all known to be less than DL, where DL is less than RL. In this report ... |
|
| Statistical Analysis of Results from the 1995 Survey of CFC Consumption on U.S. Navy Ships |
AUG 95 |
42 pages |
| Authors:
Kevin C. Burns; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| production of chlorofluorocarbon (CFC) refrigerants will cease in the United States at the end of 1995. The Naval Sea Systems Command (NAVSEA) needs to monitor total fleet requirements of CFCs until they are phased out. Stockpile requirements for ozone-depleting substances were determined from the results of a survey covering the period from 1 May 1993 to 30 April 1994. A second survey was conducted for the period from 1 May ... |
|
| Statistical Methodology for Navy Programs |
APR 90 |
14 pages |
| Authors:
Dennis E. Smith; Kevin C. Burns; DESMATICS INC STATE COLLEGE PA
|
| This research effort was composed of two tasks. The first consisted of research on statistical methodology and the implementation of statistical techniques for various problems, primarily associated with the Navy's Pollution Abatement and Ship Auxiliary Machinery programs. The second task was to investigate development of a statistical decision aid for use by Navy Program Managers in making trade-offs when allocating resources among their various programs. Section II provides a brief ... |
|
| Research on Statistical Methodology Applicable to Technical Problems Associated with Navy Programs |
OCT 87 |
|
| Authors:
Dennis E. Smith; Kevin C. Burns; DESMATICS INC STATE COLLEGE PA
|
 | This final technical report summarizes a research investigation conducted by Desmatics, Inc. under sponsorship of the Office of Naval Research. The Desmatics research effort investigated statistical methodology and developed statistical techniques to aid the Navy in reaching the major goal of the Navy's Ship Auxiliary Machinery Program. That goal is to provide equipments and procedures for existing and future ships. Keywords: Statistical Research; Data analysis; Experimental design; Statistical planning; Estimation. ... |
|
| Statistical Impact Acceleration Injury Prediction Models Based on -G sub x Accelerator Data |
OCT 86 |
|
| Authors:
Carl A. Mauro; Kevin C. Burns; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Since 1974, the Naval Biodynamics Laboratory (NBDL) has collected, as part of its research effort on acceleration impact injury prevention, an extensive data base from -G sub x accelerator runs on Rhesus subjects. Over this time period, Desmatics, Inc. has been actively involved in the development of statistically based models to predict injury in Rhesus subjects resulting from indirect head/neck impact acceleration in the -X direction. The objectives of the ... |
|
| COSTCASTER: A Cost Prediction and Trade-Off Model for Air Force Ground Communications-Electronics Equipment |
20 SEP 85 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| XCOSTCASTER is a cost-prediction and trade-off model currently under development by Desmatics, Inc. for the Air Force. The model is designed for use as a decision aid in determining whether to modify, replace, or retain items of Air Force ground communications-electronics(C-E) equipment. This paper briefly describes a micro-computer-based COSTCASTER prototype developed in Lotus 1-2-3 on a Zenith Z-100. (Author) |
|
| Prototype Implementation of COSTCASTER. A Cost-Prediction and Trade-Off Model for Air Force Ground C-E Equipment |
JUL 85 |
|
| Authors:
Gregory J. Zunic; Kevin C. Burns; Robert L. Gardner; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report describes the prototype version of COSTCASTER, a cost- prediction and trade-off model being developed by Desmatics, Inc. under Air Force Contract No. F33600-82-C-0466. COSTCASTER is intended for use as a decision aid in determining whether to modify, replace or retain Air Force ground Communications-Electronics (C-E) equipment. Specifically, the COSTCASTER computer programs developed by Desmatics are designed to provide the user with forecasts of future O&S costs at the ... |
|
| An Empirical Investigation of Several Tests for the Mean of a First- Order Autoregressive Process |
MAY 1983 |
|
| Authors:
Kevin C. Burns; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| The study of the effects of unusual environments on individuals often entails the analysis of repeated measurements taken on a single subject. Unfortunately, except under very restrictive sets of assumptions, no valid statistical techniques have been developed for such an analysis. In particular, for the first-order autoregressive process no inference procedure is currently available which enables the analyst to control the probability of making an invalid conclusion. This problem is ... |
|
| Statistical Impact Acceleration Injury Prediction Models Based on -Gx Accelerator Data and Initial Head Conditions |
MAY 1983 |
|
| Authors:
Kevin C. Burns; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Statistical impact acceleration injury prediction models are developed using data from 23 high-level -G sub X acceleration runs. These runs involve Rhesus monkeys with securely restrained torsos and unrestrained heads. The models are based on peak sled acceleration and initial head conditions. The model predictions are compared with those given in an earlier report based on different data and an estimate of Fisher's information matrix is used to evaluate the ... |
|
| Identifying Fixed Support Costs in Air Force Visibility and Management of Operating and Support Costs (VAMOSC) |
APR 1983 |
|
| Authors:
Robert L. Gardner; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report is essentially a slightly expanded transcript of a presentation given at the 50th Symposium of the Military Operations Research Society (MORs) held at the U. S. Naval Academy, Annapolis, MD, in March 1983. The material represents work in progress under Contract No. F33600-80-C-0554, with the Office of VAMOSC, HQ AFLC/LO, Wright-Patterson AFB, OH. The paper describes an approach for separating installation support costs into fixed and variable components. ... |
|
| The Effect of Environmental Change in Single-Subject Experiments |
FEB 1983 |
|
| Authors:
Kevin C. Burns; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Statistical procedures for testing the mean of a first-order autoregressive model are evaluated. Two types of test statistic are considered. One involves estimating the autocorrelation and using that estimate to transform the data. The second type of test statistic is of the form T/w, where w is a function of the estimated autocorrection and T = square root of n x the mean of Y/s. The usual estimation of the ... |
|
| Factor Screening in Simulation: Evaluation of Two Strategies on Based on Random Balance Sampling |
NOV 1982 |
|
| Authors:
Carl A. Mauro; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| In the study of large, complex computer simulation models the user is often overwhelmed by the vast number of input variables. Moreover, he or she is usually confused about how to make an effective analysis of the model without performing an excessive number of runs, which tend to be costly and time consuming. Factor screening methods, which attempt to identify the more important variables, can be extremely useful in the ... |
|
| Validation of the Cost Methodologies within the Weapon Systems Support Cost System (D160) of the Air Force VAMOSC System |
SEP 1982 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| The Air Force VAMOSC system has been mandated by DOD to aid in decision-making for modification and acquisition of weapons systems. VAMOSC data will be used in cost models that support decisions to obligate millions of dollars. Therefore, the accuracy of VAMOSC's output and the validity of its logic are paramount to the utility of the system to the user. For this reason, the Air Force Office of VAMOSC is ... |
|
| Data Integration: Combining Real-World and Simulation Data |
AUG 1982 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| In any form of scientific research or decision making, it is desirable to draw upon all relevant data which is available. Unfortunately, data derived from different sources often takes on forms which are incompatible. Consequently, much of the information not used and is thereby effectively 'lost. ' Simulation users frequently find themselves in this situation when observations have been obtained from a computer model and from the corresponding real-world situation ... |
|
| Factor Screening in Simulation: Evaluation of a Random Balance/Plackett- Burman Strategy |
JUL 1982 |
|
| Authors:
Carl A. Mauro; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Large, complex simulation models usually must be studied by performing a number of simulation runs. It is often to our advantage to invest a relatively small number of runs in a preliminary experiment aimed at screening out the most important factors. The major experimental effort and expense can then be focused on these key factors. In this report we evaluate the performance of a two-stage screening strategy that is based ... |
|
| Interruptible Supersaturated Two-Level Designs |
JUN 1982 |
|
| Authors:
James L. Rosenberger; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Interruptible designs posses a robustness against possible premature termination of an experiment. We consider such two-level designs for a first- order model and present interruptible sequences which lead to the D-optimal saturated design for four to nine factors. Premature terminations of the experiment at any stage results in a supersaturated design with minimum loss of information about the factors. The loss of these designs which is measured by the pairwise ... |
|
| Research on the Development of a Statistical Impact Acceleration Injury Prediction Model from -G(x) Accelerator Runs |
JUN 1982 |
|
| Authors:
Dennis E. Smith; David Aarons; DESMATICS INC STATE COLLEGE PA
|
| Statistical impact acceleration injury prediction models are developed for the head/neck segment from data obtained during 68 -G sub x accelerator runs. These runs involved subhuman primates (Rhesus monkeys) with securely restrained torso and unrestrained head. The data was collected by the Naval Biodynamics Laboratory (NBDL) as part of its research effort on acceleration impact injury prevention. Three classes of prediction models are constructed, one based on sled profile variables, ... |
|
| Validation of Cost Allocation Methodologies |
FEB 1982 |
|
| Authors:
Dennis E. Smith; Robert L. Gardner; DESMATICS INC STATE COLLEGE PA
|
| This report presents a discussion of the validation of the algorithms used to allocate operating and support (O&S) costs in a military cost reporting system. It also provides some general guidelines that may prove of value in validation studies concerned with similar systems. (Author) |
|
| Acceleration Measures and Motion Sickness Incidence |
NOV 1981 |
|
| Authors:
Carl A. Mauro; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| An important aspect of motion sickness research is to establish quantitative relationships between sickness incidence and various parameters of the motions that induce sickness. At present, however, only wholebody vertical sinusoidal motion has been studied to any reasonable degree. The purpose of this report is to examine the predictive utility of six different characterizations of sinusoidal motion and to investigate their possible extension to dual frequency motion. (Author) |
|
| Statistical Research on Problems Associated with Navy Fire Protection and Fuel Technology Programs |
OCT 1981 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This final technical report summarizes a study which has been devoted to statistical research in a number of problem areas associated with Navy fire protection and fuel technology programs. The report reviews the research areas considered and provides a list of informal technical notes prepared under this contract. |
|
| A Monte Carlo Study of the Use of Auxiliary Information in the Development of an Impact Acceleration Injury Prediction Model |
SEP 1981 |
|
| Authors:
Dennis E. Smith; John J. Peterson; DESMATICS INC STATE COLLEGE PA
|
| This report describes a small-scale Monte Carlo investigation of procedures for incorporating various sources of auxiliary information into an impact acceleration injury prediction model. Parameter estimates are tabulated and compared for standard and modified models. Based on the results of the investigation, the procedures appear to be helpful in reducing the mean square error of predictions. (Author) |
|
| Factor Screening in Simulation: Performance of a Two-Stage Random Balance/Plackett-Burman Procedure |
AUG 1981 |
|
| Authors:
Carl A. Mauro; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| In simulation models involving many factors (input variables), usually relatively few are important. Often it is desired to screen out these important factors using a limited number of computer runs. This paper examines the performance of a two-stage screening procedure based on random balance and Plackett-Burman designs. (Author) |
|
| Preliminary Analysis of Motion Sickness Incidence Data. |
FEB 1981 |
|
| Authors:
Carl A. Mauro; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report analyzes motion sickness data obtained from experiments involving the Office of Naval Research motion generator. Based on the analysis, a mixture of two statistical populations has been postulated as an overall model of time to first emesis. (Author) |
|
| Computational Aspects of Incorporating Auxiliary Information into an Impact Acceleration Injury Prediction Model. |
FEB 1981 |
|
| Authors:
John J. Peterson; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Auxiliary information may sometimes be used in development of a mathematical model in order to improve the estimated values of unknown parameters. This report discusses computational procedures which allow the application of commonly used nonlinear estimation programs to incorporate various sources of auxiliary information into an impact acceleration injury prediction model. (Author) |
|
| An Examination of the Performance of Two-Stage Group Screening for use in Computer Simulation. |
OCT 1980 |
|
| Authors:
Carl A. Mauro; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| When only a limited number of simulation runs are available, it is usually impossible to thoroughly investigate all of the factors under consideration. Often, though, it is anticipated that only a small subset of the original factors is important. Accordingly, it is desired to screen the factors in order to help identify those that do exert an appreciable effect on the simulation response. Two-stage group screening is one possible strategy ... |
|
| A General Statistical Approach for Using Auxiliary Information in the Development of an Impact Acceleration Injury Prediction Model. |
AUG 1980 |
|
| Authors:
John J. Peterson; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report discusses general procedures for simultaneously incorporating various sources of auxiliary information into an impact acceleration injury prediction model. The sources of auxiliary information considered are supplemental continuous empirical data and a priori knowledge in the form of model parameter estimates and constraints. (Author) |
|
| Factor Screening in Computer Simulation: Considerations in Performance Evaluation. |
JUL 1980 |
|
| Authors:
Dennis E. Smith; Carl A. Mauro; DESMATICS INC STATE COLLEGE PA
|
| Although computer simulation is a major area in which factor screening situations are frequently encountered, adequate methodology has not been developed to resolve the factor screening problem. Within the constraint of a limited number of computer runs, a decision must be made about the selection of a screening strategy that will perform efficiently and effectively. This report reviews the major classes of screening designs that have been suggested and recommends ... |
|
| Comparison of Several Estimators for the Variance of a Normal Distribution when Outliers May be Present. |
JUL 1980 |
|
| Authors:
Dennis E. Smith; Denise D. Schmoyer; DESMATICS INC STATE COLLEGE PA
|
| Although much research has been directed at dealing with outliers, particularly for a N(MU, sigma sq) distribution, little work has been devoted to estimating sigma sq when outliers may be present. This report describes a comparison of some proposed estimators of sigma sq for the case of data which, except for spurious observations, results from a N(MU sigma sq) distribution. All the estimators considered are nonadaptive and can accommodate up ... |
|
| Optimal Augmentation of Experimental Designs for Estimation of the Logistic Function. |
MAY 1980 |
|
| Authors:
Leslie A. Kalish; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| A criterion for optimal augmentation of an experimental design is applied to the problem of estimating the logistic function. A simulation study is conducted to evaluate the procedure in the two-parameter case. Examples in the development of impact acceleration injury prediction models are given. (Author) |
|
| An Investigation of a Statistical Procedure for Monitoring Two-Sample Life Tests. |
APR 1980 |
|
| Authors:
Leslie A. Kalish; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| A statistical procedure is described for comparing two lifetime distributions when the data is reviewed repeatedly over time. The procedure provides the capability of early decision while maintaining both a fixed significance level and fixed maximum length for the entire experiment. The effects which staggered entry, number of looks at the data and maximum test length have on power and expected test length are discussed. An application is made to ... |
|
| First-Order 'Interruptable' Designs. |
APR 1980 |
|
| Authors:
Dennis E. Smith; Denise D. Schmoyer; DESMATICS INC STATE COLLEGE PA
|
| Even a well-designed experiment interrupted before its intended completion may prove inefficient, posing serious data analysis problems and leaving important questions unanswered. This report addresses the construction of 'interruptable' designs for those cases where a first-order model is assumed and factors at only two levels are considered. The maximum absolute value (X'X) criterion was used in developing 'interruptable' designs for two different design strategies which attempt to limit the information ... |
|
| SCHEDULER: A Computer Program for Scheduling Administration of Performance Tests. |
NOV 1979 |
|
| Authors:
Robert L. Gardner ; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report describes SCHEDULER, a computer program developed by Desmatics, Inc. as an aid in solving a particular type of scheduling problem. The fact that SCHEDULER is implemented on a computer is only significant in terms of the speed with which a scheduling run can be made. The use of a computer performs no special magic in this particular scheduling application, nor in any other for that matter. The computer ... |
|
| Statistical Procedures for Extracting Optimal Predictor Variables for Use in an Impact Acceleration Injury Prediction Model. |
AUG 1979 |
|
| Authors:
Dennis E. Smith; John J. Peterson; DESMATICS INC STATE COLLEGE PA
|
| An empirical impact acceleration injury prediction model can be based on an underlying logistic function using information extracted from dynamic response data to define independent (predictor) variables. This report describes statistical procedures for the extraction of optimal predictor variables. The application of the statistical techniques of principal components analysis and canonical correlation analysis is described. An outline of how the data analysis may be conducted with the BMDP statistical computer ... |
|
| An Examination of Statistical Impact Acceleration Injury Prediction Models Based on Torque and Force Variables. |
JUL 1979 |
|
| Authors:
Dennis E. Smith; John J. Peterson ; DESMATICS INC STATE COLLEGE PA
|
| This technical report describes the construction of impact acceleration injury prediction models from a set of twenty-eight -G sub x accelerator runs involving Rhesus monkeys with securely restrained torso and unrestrained head. Peak torque and force variables measured during these runs were used to predict injury likelihood. The relative contribution of these variables was examined with respect to the original and a modified version of the data set. Two possible ... |
|
| Statistical Research on Problems Associated with Navy Impact Acceleration and Ship Motion Programs. |
APR 1979 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report summarizes a study which has been devoted to statistical research in a number of problem areas associated with Navy impact acceleration and ship motion programs. The report reviews research results and provides a list of technical reports and other publications prepared under this contract. |
|
| Some Bayesian Inference Procedures for Use in Developing an Impact Acceleration Injury Prediction Model. |
MAR 1979 |
|
| Authors:
John J. Peterson; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Research on the development of an impact acceleration injury prediction model based primarily on information in the data has been discussed in previous technical reports. This technical report describes Bayesian inference procedures which permit the experimenter to combine his or her prior beliefs and expert knowledge with the data observed in the experiment. (Author) |
|
| Tolerance Limits and Variables Sampling Plans: Some Power Calculations for the Normal and Lognormal Distributions. |
FEB 1979 |
|
| Authors:
Terry L. King; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| In this technical report, tables for the power of a one-sided statistical hypothesis test are presented for the situation in which the statistic is distributed as a noncentral t under both the null and alternative hypothesis. Specifically, power and sample size problems are considered for a one-sided tolerance limit and for a variables sampling plan. Examples are also included to illustrate how these tables can be used in hypothesis testing ... |
|
| Statistical Inference Procedures for a Logistic Impact Acceleration Injury Prediction Model. |
DEC 1978 |
|
| Authors:
John J. Peterson; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report describes statistical inference procedures relating to a logistic impact acceleration injury prediction model. The inference procedures discussed are interval estimation, hypothesis testing, and critical envelope estimation. Fundamentally, a critical envelope defines the independent variable combination such that the probability of injury is unlikely to exceed a prespecified value. Examples of the inference procedures discussed are given using data from a Monte Carlo simulation study. (Author) |
|
| An Examination of Statistical Impact Acceleration Injury Prediction Models Based on -Gx Accelerator Data from Subhuman Primates. |
AUG 1978 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report considers the application of an impact acceleration injury prediction model to observed data from a set of twenty-eight -G sub x accelerator runs involving subhuman primates (Rhesus monkeys) with securely restrained torso and unrestrained head. The data was collected by the Naval Aerospace Medical Research Laboratory (NAMRL) Detachment as part of its research effort on acceleration impact injury prevention. Using a common data base, two different models were ... |
|
| A Study of Estimation Accuracy When Using a Logistic Model for Prediction of Impact Acceleration Injury. |
MAR 1978 |
|
| Authors:
Dennis E. Smith; Robert L. Gardner; DESMATICS INC STATE COLLEGE PA
|
| This report addresses the topic of estimation accuracy in the development of an empirically-based logistic model for predicting impact acceleration injury. Two items of central interest are the degree of accuracy which may be expected for predictions derived from a model and the sensitivity of such predictions to sample size. A Monte Carlo simulation study was undertaken to provide information relating accuracy to sample size for selected model configurations. Two ... |
|
| Ridge Regression for Nonstandardized Models. |
OCT 1977 |
|
| Authors:
Terry L. King; Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| In the usual regression model y = Xbeta + epsilon with X of full rank and epsilon approx. N(O, sigma squared I), the ordinary least squares estimator beta = (X'X) inverse X'y has extremely large mean square error when X is an ill-conditioned matrix. This paper compares ridge estimators for beta that arise when the biasing factor (k) is applied at different stages of standardization (i.e., centering and scaling), and ... |
|
| Is Ridge Regression a Panacea. |
SEP 1977 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Consider the usual regression model y = Xbeta + epsilon where X is a matrix of full rank, beta is a vector of unknown parameters, and epsilon is a vector of random errors such that E(epsilon) = 0 and Var (epsilon) = (sigma squared)I. The procedure known as ridge regression has been offered as an alternative to ordinary least squares for estimating beta, particularly in those situations where severe multicollinearity ... |
|
| Prediction of Motion Sickness Incidence: A Statistical Examination of Three Approaches. |
AUG 1977 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report summarizes three approaches which have been suggested for predicting motion sickness incidence (MSI) for actual or simulated broadband ship motion. Two of these approaches are based on the results of MSI models developed for pure sinusoidal motion. The third approach, which uses weights derived by least squares, cannot be based directly on these models. Instead, it would be developed empirically from observed MSI in experiments involving broadband motion. ... |
|
| A Statistical Study of Grab Sampling in a Closed Environment. |
FEB 1977 |
|
| Authors:
Dennis E. Smith; Terry L. King; DESMATICS INC STATE COLLEGE PA
|
| Maintaining the quality of an enclosed environment (e.g., aboard a submarine) requires monitoring of the environment by means of sampling procedures. One general type of sampling procedure is 'grab sampling', in which atmospheric samples are obtained, essentially instantaneously, in evacuated flasks. This report describes an experiment to determine (1) whether differences exist in the performance of four grab sampling methods, and (2) whether a delay in analyzing the samples affects ... |
|
| Optimization of a Computer Simulation Response. |
NOV 1976 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| Based on results to date, the statistical techniques of Response Surface Methodology (RSM) appear to be well-adapted to use in seeking an optimum simulation response. This report summarizes the optimum-seeking problem, reviews the framework of RSM, and describes an automated RSM computer program which has been developed as an alternative to manual applications of these statistical techniques. Program interface and data preparation are discussed. In addition, easily-followed examples are presented ... |
|
| A Small-Scale Investigation of Statistics for Determining the Number of Clusters in a Data Base. |
SEP 1976 |
|
| Authors:
Dennis E. Smith; Robert L. Gardner; DESMATICS INC STATE COLLEGE PA
|
| In construction of a predictive mathematical model of impact acceleration injury, changes in evoked potential response may serve to provide important information. In attempting to determine whether a number of different types of changes exist, cluster analysis may be used. This report describes a limited Monte Carlo examination of four statistics (used in conjunction with Ward's hierarchical clustering algorithm) for identifying the number of clusters built into a simulated ten-dimensional ... |
|
| When Ridge Analysis of Response Surfaces Fails. |
AUG 1976 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| The method of ridge analysis may be used in connection with a fitted second order response surface when the objective is the location of optimum conditions. This report establishes necessary and sufficient conditions under which application of ridge analysis fails. |
|
| Research on Construction of a Statistical Model for Predicting Impact Acceleration Injury. |
FEB 1976 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| This report describes a statistical approach to development of a human impact acceleration injury prediction model based on experimental data obtained from human analogs. Also discussed in the use of preinjury criteria to provide more accurate estimates of model parameters. (Author) |
|
| Automated Response Surface Methodology in Digital Computer Simulation (U). Volume I: Program Description and User's Guide. |
SEP 1975 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| A modular computer program based on the statistical techniques of Response Surface Methodology has been developed to conduct the task of seeking optimum solutions to problems being simulated on a digital computer. The program, which may be used for constrained or unconstrained optimum-seeking in conjunction with deterministic or Monte Carlo simulations, is now available to potential users. This report volume describes the program and serves as a user's guide. A ... |
|
| Automated Response Surface Methodology in Digital Computer Simulation. Volume II. Program Flowcharts and Listings. |
SEP 1975 |
|
| Authors:
Dennis E. Smith; DESMATICS INC STATE COLLEGE PA
|
| A modular computer program based on the statistical techniques of Response Surface Methodology has been developed to conduct the task of seeking optimum solutions to problems being simulated on a digital computer. The program, which may be used for constrained or unconstrained optimum-seeking in conjunction with deterministic or Monte Carlo simulations, is now available to potential users. The first volume of this report describes the program and serves as a ... |
|
Reports by Author
