|
|
|
|
IntroductionQuantitative Decisions uses the power of mathematics and computing technology to help clients make important decisions. ContentsMost of our work has been related to environmental issues. These articles reflect that experience and our experience with the most successful techniques and technology. "Taking Risks, Reaping the Benefits" is a classic story of quantitative decision making. This is the predicament that led us to environmental statistics: how much should one pay to obtain environmental information? There is no generally applicable solution, but we have an approach that is rigorous, has been applied enough to demonstrate its success, and is non-proprietary. Anybody managing multiple industrial sites, holding a large real estate portfolio, buying or selling insurance, or routinely facing high-stakes financial decisions should be aware of how quantitative decision making works and when to use it. The "Land Use Change" article appeared in the Journal of Environmental Management (2001 v. 61) as a response to two problems encountered in the detection and analysis of land cover change. The article shows how to interpolate change in a rigorous and continuous fashion between two images. It also establishes that certain published techniques for assessing change lead to incorrect results, and suggests techniques to analyze and overcome this problem. Key words: Markov transition, land use, land cover, change analysis, transition matrix, standardizing, binomial theorem, statistical estimator. (5 pp + bibliography.) "Linear Kriging" provides insight into geostatistical interpolation by means of a simple example using three data points. It first appeared on the Web in 1999 in response to claims made in a geostatistics discussion group. The "PCBs in Pipes" article [pdf format] reviews the merits and disadvantages of following a draft EPA guidance published in 1991. It is a self-contained analysis of an interesting statistical test (a "tolerance limit") based on assuming the underlying distribution concentrations is exponential (rather than normal or lognormal, as is usually the case). An appendix develops the theory, beginning with definitions, and assesses the performance of the test in two ways. Key words: PCBs, natural gas pipeline, exponential distribution, gamma distribution, power curve, null hypothesis, tolerance limit, random sampling, wipe sampling. (25 pp + five figures + two tables.) The "Open Space" article presents an approach to community open space and land use planning using multiattribute valuation theory. The "Potential Errors in Selecting Regions for a Soils Remediation" ["Errors"] link sends you to an html version of a Powerpoint presentation delivered at the Ninth Lukacs Symposium, April 1999. It makes two points, one theoretical and the other practical (using a real example): (1) conventional techniques for identifying estimating soils contamination can be strongly biased and (2) standard statistical analyses of soils investigation data, by assuming the data were randomly obtained, therefore tend to give misleading answers. The "Fitting Data" article contains the PowerPoint slides used for a one-hour invited presentation to the U.S. EPA's workgroup studying the risk of disinfection by-products (DBPs) convened in April 1999. It presents an application of recent theoretical work in statistics to assessing whether and how to analyze a data distributions as a mixture of Normal distributions. This has great potential in environmental work, especially risk assessment. The full title of the "Air Modeling" article is Predictive Modeling of Ammonia Deposition from Large Numbers of Agricultural Sources. The abstract was peer-reviewed and accepted for the 4th International Conference on Integrating GIS and Environmental Modeling, Banff, Canada, 2-8 September 2000, where the paper was presented. The "Reprojection" article was first presented on these pages and later appeared on Directions Magazine. Its goal is to describe, as simply yet as accurately as possible, what projections and datums are and how geographic data are converted from one projection to another, one datum to another.. |
ColorRamp, Memorized Calculations, Rotate, Sample, XSect, and Tissot are trademarks of Quantitative Decisions. All other products mentioned are registered trademarks or trademarks of their respective companies.
Questions or problems regarding this web site should be directed to information
(@quantdec.com).
|
