| Optimal Mapping When Datasets are Massive |
25-Mar-2008 |
9 pages |
| Authors:
Noel A Cressie; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
 | Maps are an extremely important part of any military operation and producing timely and accurate maps is an essential planning tool. Ocean-floor or terrain data are static and spatial, while meteorological or visibility data are dynamic and spatio-temporal. Data upon which maps are based can be simultaneously massive and sparse, and they are noisy. In the presence of uncertainty due to missing data and measurement-error noise, spatial and spatio-temporal statistical ... |
|
| Detection of Daytime Arctic Clouds using MISR and MODIS Data |
MAR 2006 |
25 pages |
| Authors:
Tao Shi; Eugene E. Clothiaux; Bin Yu; Amy J. Braverman; David N. Groff; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Amongst the spectral radiances available on the Moderate Resolution Imaging Spectroradiometer (MODIS) 7 are used operationally for detection of clouds in daytime polar regions. While the information content of clouds inherent in spectral radiances is familiar, the information content of clouds contained in angular radiances (i.e., radiances emanating to space from the same object but in different directions) is not. The Multi-angle Imaging Spectroradiometer (MISR) measures angular radiances to space ... |
|
| Spatial Statistics for Command and Control |
30 NOV 2005 |
7 pages |
| Authors:
Noel A. Cressie; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| One of the key tools that a commander can use to understand and control the battlespace is a map. The danger potential of (mobile) enemy weapons is a function of the location of the weapons, their accuracy, and their firepower. In this research, we develop statistically optimal maps that show danger potential. Producing such maps requires estimates of the weapons' locations at the current time. As the weapons move, the ... |
|
| Statistical Inference for Change-of-Aperture Problems in Command and Control |
09 SEP 2002 |
6 pages |
| Authors:
Noel Cressie; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| A map is a powerful way of summarizing spatial data; in the arena of Command and Control (C2), great maps can produce knowledgable command decisions. Two approaches to statistically optimal mapping are taken. The first develops spatial multiresolution Kalman filtering of data at various apertures, and the second develops constrained optimal spatial prediction to answer nonlinear C2 questions consistently. Finally, the temporal component is introduced to allow updating of current ... |
|
| Studies in Reliability Theory and Survival Analysis and in Markov Chain Monte Carlo Methods |
DEC 1998 |
4 pages |
| Authors:
Hani Doss; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| The focus of the work has been the development of Markov chain "Monte Carlo" methods in Bayesian analysis, with emphasis on applications to survival or reliability data. We have emphasized the development of methods of dealing with analysis of sensitivity to the prior distribution. In analyzing survival data coming from reliability studies, if we are interested in estimating the distribution of the lifelength of a component, ... |
|
| Studies in Reliability Theory and Survival Analysis and in Markov Chain Monte Carlo Methods |
30 SEP 1998 |
|
| Authors:
Hani Doss; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Statistical Inference for Coherent Systems from Partial Information and Markov Chain Monte Carlo Methods |
JAN 96 |
|
| Authors:
Hani Doss; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Theory of System Reliability Demonstration, Burn-In Design, and Record Statistics |
APR 92 |
|
| Authors:
Saul Blumenthal; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
 | Estimations of component near-out time were improved by developing Poisson approximations to like distributions (shape-fit). The techniques were also successfully applied to burn-in distributions. |
|
| Research in Nonparametric Incomplete Rankings Procedures, Small Sample Asymptotics, and Statistical Decision Theory |
AUG 91 |
|
| Authors:
Douglas A. Wolfe; L. M. Berliner; Michael A. Fligner; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Burn-in with Mixed Populations |
AUG 87 |
199 pages |
| Authors:
Un-Quei W. Pan; Saul Blumenthal; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Any electronic component may be either perfect or defective. The lifetime distributions of both types are assumed known. We assume that perfect items never fail. Before being used, each production lot is tested to eliminate some of its defectives, i.e., the lot is subjected to burn-in. Here, the purpose of burn-in is to ensure with a given confidence level that an item chosen randomly from the test survivors has a ... |
|
| Software for Bayesian Analysis: Current Status and Additional Needs |
15 MAY 87 |
|
| Authors:
Prem K. Goel; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| This document makes an attempt to provide comprehensive information about the existing software for data analysis within the Bayesian paradigm. The paucity of programs seems to indicate that the Bayesian software available for widespread use is still in its infancy. We have a long way to go before a general purpose Bayesian Statistical Analysis Package is made available. Alternatives for reaching this goal quickly are presented in the concluding section. ... |
|
| The Matching Methodology: Some Statistical Properties |
15 NOV 85 |
|
| Authors:
Prem K. Goel; T. Ramalingam; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Matching is defined as the methodology of merging micro-data to create larger files of data. Matching is often done to extract statistical information which cannot be obtained from the individual files that are incomplete. Current federal statistical practice involving multivariate file- merging techniques is typically not based on a formal statistical theory. In view of this situation, a survey on matching is given. All known models for matching are presented ... |
|
| A Note on the Estimation of the Integral of f(2)(x) |
1984 |
|
| Authors:
J. C. Aubuchon; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Jackknifing Kernel Type Density Estimators |
FEB 1983 |
16 pages |
| Authors:
J. S. Rustagi; S. Dynin; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Jackknifing techniques are increasingly being applied to data analysis for bias reduction. In robust estimation, several studies have recently been published giving asymptotic properties of jackknifed estimates. Cheng has demonstrated the validity of jackknifing L-estimates under various conditions on the score function. Efron has shown that jackknife turns out to be special case of his bootstrap technique. In problems of density estimation, improvement of kernel type estimates was proposed by ... |
|
| Stronger Tests Using Heteroscedastic Anova with a Simulation of Multiple Objective Budgeting Case Study |
MAR 1982 |
|
| Authors:
Edward J. Dudewicz; W. Thomas Lin; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| The traditional analysis of variance (ANOVA) is based on the assumptions of normality, independence of the statistical errors, and equality of the variances of the errors. Studies of the robustness of the F-test have shown that the violation of normality has little effect on inferences about the means. However, the violation of independence or equality of variances can have a serious effect on inferences about the means, especially if the ... |
|
| A Simulation Analysis for Ranking and Selecting the Best Combination of Production Planning and Accounting Control Systems |
MAR 1982 |
|
| Authors:
W. Thomas Lin; Hubert J. Chen; Edward J. Dudewicz; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Frequently, the objective of computer simulation experiments of accounting and business systems is to find the best policy, procedure, or decision rule. Previous accounting or business simulation studies assumed that observations taken from each population were normally distributed with unknown means and known or equal variances. Unfortunately, only in rare cases can such assumptions be expected to hold. This paper introduces a multiple-ranking procedures, which allows for unknown and unequal ... |
|
| Uniform Improvements on the Certainty Equivalent Rule in a Statistical Control Problem |
DEC 1981 |
15 pages |
| Authors:
L. Mark Berliner; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| The problem considered arises when it is desired to choose values for design variables in a linear model so that the resulting dependent random variable will be close to a prescribed constant. A decision theoretic analysis is considered. Control procedures which dominate, in risk, the commonly used certainty equivalent rule are presented. (Author) |
|
| Complete Statistical Ranking of Populations, with Tables and Applications |
DEC 1981 |
|
| Authors:
Jan Beirlant; Edward J. Dudewicz; Edward C. van der Meulen; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| A new statistical methodology called ranking and selection theory is introduced. This document reviews procedures for completely ranking a set of populations (from best, second best, etc., down to worst), gives new tables needed to implement these procedures, and considers several practical examples using real data. |
|
| A Simulation of Multiple Objective Budgeting Models Using Heteroscedastic ANOVA |
NOV 1981 |
20 pages |
| Authors:
Edward J. Dudewicz; W. Thomas Lin; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| An important problem confronting managers and accountants is design of computer simulation experiments in settings where the assumptions usually made by statisticians are violated. One approach to this problem is to run the simulation 'until the budgeted money runs out' and hope for the best analysis of the simulated results. The present paper shows how, instead, to rationally design and analyze a computer simulation under minimal assumptions, using the Bishop-Dudewicz ... |
|
| Ranking and Selection in Designed Experiments: Complete Factorial Experiments |
OCT 1981 |
25 pages |
| Authors:
Edward J. Dudewicz; Baldeo K. Taneja; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Over the years, while it has been recognized that developments for other designs are of great interest, almost all ranking an selection papers have dealt explicitly only with the completely randomized design. We review what is known in other design settings, then proceed to new results on the complete factorial experimental design. We show that existing procedures are vitiated by interaction in such a design, but give a new procedure ... |
|
| Optimization in Statistics - An Introduction |
OCT 1981 |
21 pages |
| Authors:
Jagdish S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Optimization techniques play an important role in various areas of statistics. Examples from Estimation, nonparametric statistics, design of experiments, sample surveys, multivariate statistical inference, information theory and regression analysis are provided. (Author) |
|
| Selected Bibliography on Optimizing Techniques in Statistics |
AUG 1981 |
19 pages |
| Authors:
J. S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| A selected list of references is provided in the area of optimization techniques as applied in statistical settings. The bibliography is classified according to various optimization techniques as well as to the kinds of statistical applications. Statistical areas include design of experiments, estimation and tests of hypotheses, least-squares theory, regression analysis, multivariate analysis, sampling, sequential and search procedures, and stochastic approximation methods. (Author) |
|
| Optimizing Methods in Simulation |
AUG 1981 |
44 pages |
| Authors:
J. S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Simulation techniques utilize methods of optimization in several aspects. Validation of simulation models, estimation of models and design of simulation experiments require optimization. Methods of optimization are discussed with special applications to simulation. Criteria of optimization, classical methods of optimization, numerical methods, and optimal search procedures are discussed. (Author) |
|
| Modern Design of Simulation Experiments |
JUN 1981 |
23 pages |
| Authors:
Edward J. Dudewicz; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Design of simulation experiments has long been neglected by experimenters, who tend to attempt either to run only k experiments to evaluate k factors, or to run 2**k experiments (often infeasible, always costly). In this paper we survey the possibilities and show how to design for valid inferences with practical-sized experiments. (Author) |
|
| Probabilistic Structures of Modern Lottery Games |
MAY 1981 |
37 pages |
| Authors:
Jagdish S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| An introduction to the basic structure of lottery games is given. Several examples from modern state lotteries are included. Statistical considerations such as tests of randomness as required in the conduct of lottery games are discussed. (Author) |
|
| Some Asymptotic Results for the Induced Selection Differential |
26 MAR 1981 |
|
| Authors:
H. N. Nagaraja; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Some Nondegenerate Limit Laws for the Selection Differential |
FEB 1981 |
|
| Authors:
H. N. Nagaraja; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Statistical Considerations in Control Models with Applications |
JUL 1980 |
30 pages |
| Authors:
Jagdish S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Many applications of control theory, specially in areas other than engineering, require the estimation of the system equations from available data. Applications from patient care system, effect of air pollution on health and dynamic economic models are discussed. A framework is studied in which the control model can be applied using the data provided by the system itself. Special objective functions are utilized to illustrate the technique. (Author) |
|
| Estimation of the Larger Mean |
JUL 1980 |
23 pages |
| Authors:
Ishwari D. Dhariyal; Edward J. Dudewicz; Saul Blumenthal; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Estimation of the larger of two normal means mu 1 and mu 2 is studied. Two new estimators are studied for this problem, the maximum probability estimator and the iterated bias elimination estimators. If abs. val. mu 1 - mu 2 is not close to zero (as evidenced by a strongly consistent estimator), a maximum probability estimator is seen to be best as regards both bias and mean-squared error. (Author) ... |
|
| Estimation of the P(CS) with a Computer Program. Revision |
JUL 1980 |
|
| Authors:
Edward J. Dudewicz; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| In the area of ranking and selection, recent papers have presented procedures which attempt to allow experimenters to take at least partial advantage of more-favorable configurations of the population parameters (compared to least-favorable configuration). The current paper presents a computer program which allows experimenters to dispense with all of inequalities, an indifference-zone, and known equal variabilities in their assessment of the probability of correct selection. It has additional uses in ... |
|
| Heteroscedasticity |
JUL 1980 |
28 pages |
| Authors:
Edward J. Dudewicz; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Recently developed methods of dealing with heteroscedasticity without transformations, in ANOVA, confidence intervals, and ranking and selection, are given. Practical aspects, a numerical example, contrasts with other approaches, and critical analysis are given. (Author) |
|
| A Multivariate Solution of the Multivariate Ranking and Selection Problem |
FEB 1980 |
|
| Authors:
Edward J. Dudewicz; Vidya S. Taneja; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| The problem of selection of the best multivariate population is given a new formulation which does not involve reducing the populations to univariate quantities. This formulation's solution is developed for known, and (using the Heteroscedastic Method) also for unknown, variance-covariance matrices. Preference reversals and arbitrary nonlinear preference functions are explicity allowed in this new theory. (Author) |
|
| Analysis of Variance with Unequal Variances |
NOV 1979 |
|
| Authors:
Edward J. Dudewicz; Thomas A. Bishop; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Analysis of variance (ANOVA) is often used in quality control studies. It assumes equal variabilities within groups, and no exact procedures have been available for cases with unequal variabilities. In this paper exact procedures are given and illustrated. An indication of the losses to be incurred by using the traditional F-test when variances are unequal is given. (Author) |
|
| Recent Applications of Variational Techniques in Statistics |
AUG 1979 |
17 pages |
| Authors:
J. S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Recent applications of variational methods to problems of statistical inference are given. The problems of robustness, inadmissibility and penalized maximum likelihood estimates are discussed to exhibit their variational character. (Author) |
|
| Estimation of Rate Constants in the Michaelis - Menten Model |
MAR 1979 |
18 pages |
| Authors:
J. S. Rustagi; Joanne Yang; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Nonlinear models leading to Michaelis-Menten scheme result from the study compartment systems as well as in other contexts. Using difference equation approach, rate constants are estimated in some simple models. Simulation methods are used to derive the distributions of these estimates. It is shown that the assumptions of normality of errors leads to the normal distribution of the estimated rate constants. (Author) |
|
| Approximate Tests of Hypotheses in Regression Models with Grouped Data |
FEB 1979 |
|
| Authors:
Abhaya Indrayan; Jagdish S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Techniques for testing hypotheses about parameters in the regression models under the situation of grouped data are provided. A test statistic similar to conventional F statistic is considered. A simulation study performed for a few cases shows that the proposed statistic has an approximate F distribution and is useful in applications. (Author) |
|
| A Multivariate Solution of the Multivariate Ranking and Selection Problem |
JUL 1978 |
|
| Authors:
Edward J. Dudewicz; Vidya S. Taneja; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
|
| Approximate Maximum Likelihood Estimates in Regression Models for Grouped Data |
1978 |
23 pages |
| Authors:
A. Indrayan; J. S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| Grouped data arise quite frequently in application. Various experimental situations naturally lead to observations classified into certain groups. The process of recording or storing observations directly leads to grouped data. Other examples of circumstances which lead to grouped data have recently been discussed by Haitovsky (1973) in his monograph on regression estimation from grouped observations. From a practical point of view, the case for grouped data has been advocated by ... |
|
| Statistics and Probability Applied to Control System Analysis |
31 AUG 1973 |
|
| Authors:
J. S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| The research topics discussed in the paper include: Variational methods in statistics, inference in Markov dependent firing distribution; paired comparisons; sampling theory and its applications to the development of new sampling strategy for monitoring air pollutants; and, Pareto distribution. |
|
| Studies in Statistics and Applied Probability with Applications |
SEP 1972 |
31 pages |
| Authors:
J. S. Rustagi; R. C. Srivastava; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| The research in problems in applied probability and statistics during the five years of the project was concerned with waiting time distribution of Markov dependent Bernoulli trials, variational methods in statistics, stochastic processes and miscellaneous problems. (Author) |
|
| A Note on Homogeneous Processes with Independent Increments |
JUN 1972 |
|
| Authors:
Y. H. Wang; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| A class of stochastic processes, X(t) is characterized based in the property that the conditional mean and variance of X(t), given X(t sub 1) = y, for some 0 < t < (t sub 1), are linear functions of y. Two particular cases resulting in Wiener and Poisson processes are discussed. (Author) |
|
| Symposium on Optimizing Methods in Statistics |
31 OCT 1971 |
|
| Authors:
Jagdish S. Rustagi; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| A symposium on optimizing methods in statistics was held at The Ohio State University during June 14-16, 1971. The program of the sessions included the following topics: Variational methods; Regression analysis; Optimum seeking methods; Mathematical programming; Stochastic control; Optimum design of experiments; Optimum spacing and order statistics; problems and applications; and Contributed papers. |
|
| A Stochastic Model for Air Interdiction |
JUN 1971 |
|
| Authors:
D. E. Crawford; R. C. Srivastava; OHIO STATE UNIV COLUMBUS DEPT OF STATISTICS
|
| In a recent paper, Armitage (Allocation of Sorties in Air Interdiction, Journal of the Operations Research Society of America, XVIII, No. 3 (May-June), pp. 483-496) has considered a deterministic model of air interdiction in a simplified hostile country through which material is supplied. In this report, an analogous stochastic model is developed. Three important elements of a stochastic model for air interdiction are stochastic processes describing the movement of trucks ... |
|
Reports by Corporate Author
