Ranking of Functions

Selecting the best distribution function is not a trivial task. A wide variety of statistical data can be used in this duty, including standard deviations, R, Akaike and Bayesian Information Criteria, and even CPU time, which are aU presented in Table 12.23. It is well accepted that correlation coefficients are not very useful in discriminating between models. In this study, the correlation coefficients were very close to unity (0.986-0.999) for all of the functions. To highlight this point, only the Alpha distribution function exhibited a value of R lower than 0.99. 

One method to rank the models is to compare the standard deviations. Since, the standard deviation values for all functions ranged from 0.59 to 3.37 wt%, they were more useful than the R values or the slopes and intercepts from the parity plots. 

CPU-time relative to that required with normal distribution function. 

FIGURE 12.14 Residual plot for (a) normalized alpha and (b) WeibuU extreme distribution functions. 

From the results given in Table 12.23 (R, slope, intercept, and SD) and residual analysis, the following classification of accuracy of predictions was established  

---

Examination of Fig. 5 shows how all these different pieces of information are correlated to generate a numerical ranking of functional product requirement importance. Clearly, QFD is at best a semi quantitative tool, particularly at the top level. CTQ importance... 

Beyond presenting hypotheses and concepts, one also obtains initial measurements of validty, acceptability, and viability. These measurements are in terms of quantitative ratings and rankings of functions and features, as weU as more free-flow comments and dialogue. 

Order of precedence of functions (Section 16.2B) A ranking of functional groups in order of priority for the purposes of lUPAC nomendature. 

A collection of databases of chemicals and of functional groups which rank chemicals and groups relative to their reactivity, stability, toxicity, and flammability categories. This would assist in the evaluation of the potential benefits of substituting one, somewhat safer, chemical for another. 

The functions of the Academie Royal des Sciences were assumed in 1795 by a branch of the newly formed National Institute. Laplace was elected vice president of this reincarnated Academy and then elected president a few months later, in 1796. The duties of this position put him in contact with Napoleon Bonaparte. Three weeks after Napoleon seized power m 1799, Laplace presented him with copies of his work on celestial mechanics. Bonaparte quipped that he would read it in the first six weeks I have free and invited Laplace and his wife to dinner. Three weeks later, Napoleon named Laplace his minister of the interior. After six weeks, however, he was replaced Napoleon thought him a complete failure as an administrator. However, Napoleon continued to heap honors and rewards upon him, regarding him as a decoration of the state. lie made Laplace a chancellor of the Senate with a salai y that made him wealthy, named him to the Legion of Honor, and raised him to the rank of count of the empire. Laplace s wife was appointed a lady-in-waitmg to the Italian court of Napoleon s sister. Laplace responded with adulatory dedications of his works to Napoleon. 

In the next step, the rank is calculated of the difference matrix X = X - kX. For any value of k, the rank of X is equal to 1 + n, except for the case where k is exactly equal to the contribution of the analyte to the signal. In this case the rank of X is / - 1. Thus the concentration of the analyte in the unknown sample can be found by determining the k-value for which the rank of Xj is equal to / - 1. The amount of the analyte in the sample is then equal to kc where is the concentration of the analyte in the standard solution. In order to find this k-value Ho et al. proposed an iterative procedure which plots the eigenvalues of the least significant PC of X as a function of k. This eigenvalue becomes minimal when k exactly compensates the signal of the analyte in the sample. For other k-values the signal is under- or... 

To illustrate this latter point, Figure 5 shows the calculated amount of coal required for hydrogen manufacture as a function of the rank of the starting coal and the composition of the desired products (10). In these calculations a 12.5% methane byproduct was assumed and the thermal efficiency of the hydrogen generation was assumed to be 70%. 

It has already been shown that the Cone calorimeter smoke parameter correlates well with the obscuration in full-scale fires (Equation 1). At least four other correlations have also been found for Cone data (a) peak specific extinction area results parallel those of furniture calorimeter work [12] (b) specific extinction area of simple fuels burnt in the cone calorimeter correlates well with the value at a much larger scale, at similar fuel burning rates [15] (c)maximum rate of heat release values predicted from Cone data tie in well with corresponding full scale room furniture fire results [16] and (d) a function based on total heat release and time to ignition accurately predicts the relative rankings of wall lining materials in terms of times to flashover in a full room [22]. 

The generic rank of a structured matrix B is defined to be the maximal rank that B achieves as a function of its free parameters. 

As was indicated in Section 7.2, the vector of measurement adjustments, e, has a multivariate normal distribution with zero mean and covariance matrix V. Thus, the objective function value of the least square estimation problem (7.21), ofv = eT l> 1 e, has a central chi-square distribution with a number of degrees of freedom equal to the rank of A. 

Chemical reactions for which the rank of the reaction coefficient matrix T is equal to the number of reaction rate functions R, (i. e 1,..., I) (i.e., Nr = I), can be expressed in terms of / reaction-progress variables Y, (i. e 1,...,/), in addition to the mixture-fraction vector . For these reactions, the chemical source terms for the reaction-progress variables can be found without resorting to SVD of T. Thus, in this sense, such chemical reactions are simple compared with the general case presented in Section 5.1. 

However, care must be taken to avoid the singularity that occurs when C is not full rank. In general, the rank of C will be equal to the number of random variables needed to define the joint PDF. Likewise, its rank deficiency will be equal to the number of random variables that can be expressed as linear functions of other random variables. Thus, the covariance matrix can be used to decompose the composition vector into its linearly independent and linearly dependent components. The joint PDF of the linearly independent components can then be approximated by (5.332). 

Telechelic polymers rank among the oldest designed precursors. The position of reactive groups at the ends of a sequence of repeating units makes it possible to incorporate various chemical structures into the network (polyether, polyester, polyamide, aliphatic, cycloaliphatic or aromatic hydrocarbon, etc.). The cross-linking density can be controlled by the length of precursor chain and functionality of the crosslinker, by molar ratio of functional groups, or by addition of a monofunctional component. Formation of elastically inactive loops is usually weak. Typical polyurethane systems composed of a macromolecular triol and a diisocyanate are statistically simple and when different theories listed above are... 

Within the Matlab s numerical precision X is singular, i.e. the two rows (and columns) are identical, and this represents the simplest form of linear dependence. In this context, it is convenient to introduce the rank of a matrix as the number of linearly independent rows (and columns). If the rank of a square matrix is less than its dimensions then the matrix is call rank-deficient and singular. In the latter example, rank(X)=l, and less than the dimensions of X. Thus, matrix inversion is impossible due to singularity, while, in the former example, matrix X must have had full rank. Matlab provides the function rank in order to test for the rank of a matrix. For more information on this topic see Chapter 2.2, Solving Systems of Linear Equations, the Matlab manuals or any textbook on linear algebra. 

PLS regression as described in Section 4.7 allows finding (linear) relations between two data matrices X and Y that were measured on the same objects. This is also the goal of CCA, but the linear relations are determined by using a different objective function. While the objective in the related method PLS2 is to maximize the covar lance between the scores of the x- and y-data, the objective of CCA is to maximize their correlation. In CCA, it is usually assumed that the number n of objects is larger than the rank of X and of Y. The reason is that the inverse of the covariance matrices are needed which would otherwise not be computable applicability of CCA to typical chemistry data is therefore limited. 

---