Reality statistical

Throughout this book we have emphasized fundamental concepts, and looking at the statistical basis for the phenomena we consider is the way this point of view is maintained in this chapter. All theories are based on models which only approximate the physical reality. To the extent that a model is successful, however, it represents at least some features of the actual system in a manageable way. This makes the study of such models valuable, even if the fully developed theory falls short of perfect success in quantitatively describing nature. 

Practical considerations usually limit the number of replicate specimens of each kind that can be exposed for each period of test. At least two are recommended for obvious reasons, and if a larger number can be accommodated in the programme more valuable results can be secured—especially when it is desired to establish the reality of small differences in performance. For statistical analysis, five replicates are desirable. Accounts of statistical planning and analysis are given by F. H. Haynie in Reference 2 and in ASTM 016 1984. 

FIGURE 2.2. A schematic description of the evaluation of the transmission factor F. The figure describes three trajectories that reach the transition state region (in reality we will need many more trajectories for meaningful statistics). Two of our trajectories continue to the product region XP, while one trajectory crosses the line where X = X (the dashed line) but then bounces back to the reactants region XR. Thus, the transmission factor for this case is 2/3. 

Statistical tests incorporate mathematical models against which reality, perhaps unintentionally or unwillingly, is compared, for example ... 

In the past three decades, industrial polymerization research and development aimed at controlling average polymer properties such as molecular weight averages, melt flow index and copolymer composition. These properties were modeled using either first principle models or empirical models represented by differential equations or statistical model equations. However, recent advances in polymerization chemistry, polymerization catalysis, polymer characterization techniques, and computational tools are making the molecular level design and control of polymer microstructure a reality. 

In the previous Maxwelhan description of X-ray diffraction, the electron number density n(r, t) was considered to be a known function of r,t. In reality, this density is modulated by the laser excitation and is not known a priori. However, it can be determined using methods of statistical mechanics of nonlinear optical processes, similar to those used in time-resolved optical spectroscopy [4]. The laser-generated electric field can be expressed as E(r, t) = Eoo(0 exp(/(qQr ot)), where flo is the optical frequency and q the corresponding wavevector. The calculation can be sketched as follows. 

Perhaps the most interesting aspect of this set of studies is the question posed in the recent paper by Schmidt et al. (2004) and deals with the reality of the patterns they observed. Is the polymorphism observed a result of the calculation methods used in the study, neural network (NN), and multivariate statistical analysis (MVA) Would increased sampling result in a greater number of chemo-types It is entirely possible, of course, that the numbers obtained in this study are a true reflection of the biosynthetic capacities of the plants studied. The authors concluded—and this is a point made elsewhere in this review—that ... for a correct interpretation a good knowledge of the biosynthetic background of the components is needed. ... 

The consequence of all these (conscious and unconscious) simplifications and eliminations might be that some information not present in the process will be included in the model. Conversely, some phenomena occurring in reality are not accounted for in the model. The adjustable parameters in such simplified models will compensate for inadequacy of the model and will not be the true physical coefficients. Accordingly, the usefulness of the model will be limited and risk at scale-up will not be completely eliminated. In general, in mathematical modelling of chemical processes two principles should always be kept in mind. The first was formulated by G.E.P. Box of Wisconsin All models are wrong, some of them are useful . As far as the choice of the best of wrong models is concerned, words of S.M. Wheeler of New York are worthwhile to keep in mind The best model is the simplest one that works . This is usually the model that fits the experimental data well in the statistical sense and contains the smallest number of parameters. The problem at scale-up, however, is that we do not know which of the models works in a full-scale unit until a plant is on stream. 

The alternative to compartmental analysis is statistical moment analysis. We have already indicated that the results of this approach strongly depend on the accuracy of the measurement process, especially for the estimation of the higher order moments. In view of the limitations of both methods, compartmental and statistical, it is recommended that both approaches be applied in parallel, whenever possible. Each method may contribute information that is not provided by the other. The result of compartmental analysis may fit closely to the data using a model that is inadequate [12]. Statistical moment theory may provide a model which is closer to reality, although being less accurate. The latter point has been made in paradigmatic form by Thom [13] and is represented in Fig. 39.16. 

Transition state theory, as embodied in Eq. 10.3, or implicitly in Arrhenius theory, is inherently semiclassical. Quantum mechanics plays a role only in consideration of the quantized nature of molecular vibrations, etc., in a statistical fashion. But, a critical assumption is that only those molecules with energies exceeding that of the transition state barrier may undergo reaction. In reality, however, the quantum nature of the nuclei themselves permits reaction by some fraction of molecules possessing less than the energy required to surmount the barrier. This phenomenon forms the basis for QMT. ... 

The relation of measured results to given values, e.g., critical levels, legally fixed values, regulatory limits, maximum acceptable values, is of continual relevance in analytical chemistry. In the analytical reality, the problematic nature of detection leads to the test statistics, strictly speaking to the t-test (Currie [1995, 1997] Ehrlich and Danzer [2006]). By means of that, it is tested, if the determined analytical result is significantly different from the average blank of the critical value, respectively. 

Particle shape plays an important role in particle size determination. The simplest definition of particle size diameter is based on a sphere, which has a unique diameter. In reality, however, many particles are not well represented by this model. Figure 1 illustrates the variety of shapes that may be found in particle samples [1]. As the size of a particle increases, so does its tendency to have an irregular shape [2], complicating statistical analysis. Particle shape coefficients have been derived for different geometries [3], and various equivalent diame-... 

As mentioned earlier, a random and statistical cyclization with two different and regio-defined Ti-compounds would produce a synthetically unattractive mixture of ten different zirconacycles. In reality, however, there are a few factors that can be exploited to produce a single desired zirconacycle. A systematic investigation has revealed that there are several discrete types offive-membered zirconacycle formation, as shown in Scheme 1.53 [88,89] (Generalization 20). In the Type I reaction, the cross-selective cyclization is kinetically favored. Presumably, little ethylene is displaced during the reaction. Type I reactions cannot be readily observed with ZrCp2 complexes with 1-butene. In contrast, Type II cyclization must be thermodynamically controlled, as 1-butene is readily displaced by a number of better Ti-ligands. It is predicted, however, that the cross-combination of the two Ti-com-... 

The method detection limit is, in reality, a statistical concept that is applicable only in trace analysis of certain types of substances, such as organic pollutants by gas chromatographic methods. The method detection limit measures the minimum detection limit of the method and involves all analytical steps, including sample extraction, concentration, and determination by an analytical instrument. Unlike the instrument detection limit, the method detection limit is not confined only to the detection limit of the instrument. 

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. 

The question that must be asked is, given the total number of animals actually assayed was relatively small in comparison with the total number of animals produced, how significant are the results Based on the statistical design and the observation that the range (95% confidence limits) will reflect closely the realities of the animal population sampled, the estimate of the frequency of residues in any given slaughter class is a good estimate. 

The reality of HTS is that the true statistical hit rate is often greater than the operational hit rate that can be accommodated by confirmation assays. In such a case, the use of a Top X method carries a number of significant drawbacks, the most significant of which is the creation of artificially high false negative rates through the neglect of actives less potent than the cutoff used. This is often overlooked but is of critical importance. 

The statistical significance relates strictly to the conditions under which the trial was conducted and will tell how often a difference of the observed size could occur by chance alone if there is, in reality, no difference between the treatments. The most widely accepted level of probability in therapeutic trials is set at 5%,... 

How can statistical error be estimated for a single observable from independent simulations There seems little choice but to calculate the standard error in the mean values estimated from each simulation using Equation (1), where the variance is computed among the averages from the independent simulations and NJ L> is set to the number of simulations. In essence, each simulation is treated as a single measurement, and presumed to be totally independent of the other trajectories. Importantly, one can perform a "reality check" on such a calculation because the variance of the observable can also be calculated from all data from all simulations — rather than from the simulation means. The squared ratio of this absolute variance to the variance of the means yields a separate (albeit crude) estimate of the number of independent samples. This latter estimate should be of the same order as, or greater than, the number of... 

Repeating a routine analysis over and over again for a period of time (perhaps sometimes years) and assembling the results into a data set that is free of bias and determinate errors create a basis for calculating a standard deviation that approaches o, the true standard deviation. The 2o theoretically associated with 95.5% of the values (Section 4.3), or the 3o associated with 99.7% of the values then comes close to reality. If a given analysis result on a given day is then within 2o, it is a signal that "all is well" and the process or procedure is considered to be under what is called statistical control. If a process or procedure is under statistical control, then only 4.5% of the points (about 1 of every 20) would be outside the 2a limits and only... 

A probabilistic risk assessment (PRA) deals with many types of uncertainties. In addition to the uncertainties associated with the model itself and model input, there is also the meta-uncertainty about whether the entire PRA process has been performed properly. Employment of sophisticated mathematical and statistical methods may easily convey the false impression of accuracy, especially when numerical results are presented with a high number of significant figures. But those who produce PR As, and those who evaluate them, should exert caution there are many possible pitfalls, traps, and potential swindles that can arise. Because of the potential for generating seemingly correct results that are far from the intended model of reality, it is imperative that the PRA practitioner carefully evaluates not only model input data but also the assumptions used in the PRA, the model itself, and the calculations inherent within the model. This chapter presents information on performing PRA in a manner that will minimize the introduction of errors associated with the PRA process. 

The mathematical obfuscation of these models must not remove the requirement that every receptor model must be representative of and derivable from physical reality as represented by the source model. A statistical relationship between the variability of one observable and another is Insufficient to define cause and effect unless this physical significance can be established. 

---