Statistics and statisticians

Lewis JA (1996) Statistics and statisticians in the regulation of medicines. Journal of the Royal Statistical Society Series A - Statistics in Society 159 359-362. 

C. Daniel, App/ications of Statistics to lndustria/Experimentation, ]oE Wiley Sons, Inc., New York, 1976. This book is based on the personal experiences and insights of the author, an eminent practitioner of industrial appHcations of experimental design. It provides extensive discussions and concepts, especially in the areas of factorial and fractional factorial designs. "The book should be of use to experimenters who have some knowledge of elementary statistics and to statisticians who want simple explanations, detailed examples, and a documentation of the variety of outcomes that may be encountered." Some of the unusual features are chapters on "Sequences of fractional repHcates" and "Trend-robust plans," and sections entided, "What is the answer (what is the question )," and "Conclusions and apologies."... 

Chemometricians do not believe that good calibration diagnostics properly interpreted can estimate prediction performance, and insist on a separate validation data set. Statisticians, on the other hand, do believe that. Certainly, it is good practice and statisticians also prefer to verify the estimates through the use of validation data when that is available, but in some cases they are not always available. In those cases, having generalized statistics available so that you can know when a model will be a good estimate of prediction performance is a major benefit. 

Conover, J.W. and Inman, R.L. (1981). Rank transformation as a bridge between parametric and nonparametric statistics. Am. Statistician 35 124-129. 

Over the years people have switched their beliefs from the explanations of the gods of former times to the explanations of the gods of modern times, the scientists and statisticians It Is easy to see how people have done this Modern people began to believe In numbers and data because they represented nature so well. The theories of science and statistics were explained, and the Ideas of science and statistics became better founded and entrenched In their thinking. However one mistake was made and that was a big one facts and data were taken as truth and reality. 

Once the design is established, a protocol will be prepared. All the critical design features and the types of observations to be made, and even the statistical methods to be used to analyze results, are specified in advance. Toxicologists, chemists, pathologists, and statisticians are typically involved in drafting the protocol. 

The fall of 1983 also saw the North Atlantic Treaty Organization host an Advanced Studies Institute in Cosenza, Italy, entitled Chemometrics Mathematics and Statistics in Chemistry. One hundred scientists—a most unusual collection of chemists, engineers, and statisticians from academia, industry, and government—representing a dozen countries assembled to discuss the role of sophisticated multivariate statistics in the daily routine of an analytical chemistry laboratory. 

The likely effectiveness of the product, derived from the preclinical and early clinical work, will determine study design, complexity and size. It is a mistake to try to answer too many questions in a single study, despite the apparent commercial attractiveness of such a strategy. A study overburdened by many secondary objectives is more likely to fail when the design is implemented in many centres worldwide. What seems a good idea in head office can often be hard to implement in the clinic. Statistical advice is vital, and statisticians offer excellent opinions about the utility of complex study designs. 

In various sections of this paper, the authors appropriately emphasize the needed professional knowledge, experience, integrity and responsibility of the analyst. The handling of samples, the estimation of uncertainties, and the vigilance for unexpected errors also require some familiarity with statistics and possibly the help of a statistician as a consultant. However, the final assignment of uncertainties is the responsibility of the analyst who has actually performed the analyses. 

One technique employed to arrive, at an appropriate value has been postulated by L. Torbeck [16], who has taken a statistical and practical approach that in the absence of any other retest rationale can be judged as a technically sound plan. According to Torbeck, the question to be answered— how big should the sample be —is not easily resolved. One answer is that we first need a prior estimate of the inherent variability, the variance, under the same conditions to be used in the investigation. What is needed is an estimate of a risk level (defined as the percentage of probability that a significant difference exists between samples when there is none what statisticians call a type 1 error), the (i risk level ([1 is the probability of concluding that there is no difference between two means when there is one also known as a type 2 error) and the size of the difference (between the OOS result and the limit value) to be detected. The formula for the sample size for a difference from the mean is expressed as ... 

The interpretation of the results of all trials requires the close collaboration of clinicians and statisticians. Reliable results are only obtained if at least the minimum number of patients for statistical viability are involved in the preliminary trials. It is often difficult to measure precisely the parameter chosen for assessment. Consequently, results are usually quoted in terms of a probability coefficient, the lower the value of this coefficient the more accurate the results. However, very reliable results will only be obtained from clinical trials if large groups of patients are tested. This is seldom feasible. Consequently, manufacturers and licencing authorities usually settle for the best statistical compromise. Since some adverse effects do not manifest themselves for years, it is necessary to constantly monitor the drug (Phase IV trials) after it has been released for general use. 

Chemometrics is an interdisciplinary field that combines statistics and chemistry. From its earliest days, chemometrics has always been a practically oriented subdiscipline of analytical chemistry aimed at solving problems often overlooked by mainstream statisticians. An important example is solving multivariate calibration problems at reduced rank. The method of partial least-squares (PLS) was quickly recognized and embraced by the chemistry community long before many practitioners in statistics considered it worthy of a second look. ... 

Loyer, M. (1983). Bad probability, good statistics, and group testing for binomial estimation. The American Statistician, 37, 57-59. 

Chemists and statisticians use the term mixture in different ways. To a chemist, any combination of several substances is a mixture. In more formal statistical terms, however, a mixture involves a series of factors whose total is a constant sum this property is often called closure and will be discussed in completely different contexts in the area of scaling data prior to principal components analysis (Chapter 4, Section 4.3.6.5 and Chapter 6, Section 6.2.3.1). Hence in statistics (and chemometrics) a solvent system in HPLC or a blend of components in products such as paints, drugs or food is considered a mixture, as each component can be expressed as a proportion and the total adds up to 1 or 100%. The response could be a chromatographic separation, the taste of a foodstuff or physical properties of a manufactured material. Often the aim of experimentation is to find an optimum blend of components that tastes best, or provide die best chromatographic separation, or die material diat is most durable. 

NIRS involves the multidisciplinary approaches of the analytical chemist, statistician, and computer programmer. The word chemometrics refers to the application of mathematical or statistical methods to measurements made on chemical systems of varying complexity. Chemometrics is defined as the chemical discipline that uses mathematical, statistical, and other methods that apply formal logic to design or select optimal measurement procedures and experiments, and to provide maximum relevant chemical information by analyzing chemical data. 

Does not collect data, but maintains a collection of statistical publications, a bibliographic program and reference service, and a directory of Latin American and international statistical agencies and statisticians. Office of the Coordinator of International Statistics, Bureau of the Census, Washington 25, D. C. ... 

---