Building validation

An important aspect of all methods to be discussed concerns the choice of the model complexity, i.e., choosing the right number of factors. This is especially relevant if the relations are developed for predictive purposes. Building validated predictive models for quantitative relations based on multiple predictors is known as multivariate calibration. The latter subject is of such importance in chemo-metrics that it will be treated separately in the next chapter (Chapter 36). The techniques considered in this chapter comprise Procrustes analysis (Section 35.2), canonical correlation analysis (Section 35.3), multivariate linear regression... 

Harm, B. and Balmain, A. (2001) Building validated mouse models of human cancer. Curr Opin Cell Biol 13, 778-784. 

Without empirically based health economic evaluations to guide an understanding of allergy and asthma prevention strategies, it would seem that the next major step will rest on building valid, reliable, and dynamic health economic models derived from the larger body of literature based on the tertiary control of asthma and allergic rhinitis. 

Figure 10.3 Large observed sample split into three sets training for model building, validation for model selection, and test for assessment of prediction accuracy. (See color insert.)...

Another method of detection of overfitting/overtraining is cross-validation. Here, test sets are compiled at run-time, i.e., some predefined number, n, of the compounds is removed, the rest are used to build a model, and the objects that have been removed serve as a test set. Usually, the procedure is repeated several times. The number of iterations, m, is also predefined. The most popular values set for n and m are, respectively, 1 and N, where N is the number of the objects in the primary dataset. This is called one-leave-out cross-validation. 

Iteration of the steps, descriptor selection, model building, and model validation in combination with an optimi ation algorithm allows one to select a descriptor subset having maximum predictivity. 

The abbreviation QSAR stands for quantitative structure-activity relationships. QSPR means quantitative structure-property relationships. As the properties of an organic compound usually cannot be predicted directly from its molecular structure, an indirect approach Is used to overcome this problem. In the first step numerical descriptors encoding information about the molecular structure are calculated for a set of compounds. Secondly, statistical methods and artificial neural network models are used to predict the property or activity of interest, based on these descriptors or a suitable subset. A typical QSAR/QSPR study comprises the following steps structure entry or start from an existing structure database), descriptor calculation, descriptor selection, model building, model validation. 

Another way to examine scaleup of hydrodynamics is to build a cold or hot scale model of the commercial design. Validated scaling criteria have been developed and are particularly effec tive for group B and D materials [Gheksmau, Hyre and Woloshuu, Powder Tech., 177-199 (1993)]. 

This model of Cro binding to DNA was arrived at by intuition and clever model building. Its validity was considerably strengthened when the same features were subsequently found in the DNA-binding domains of the lambda-repressor molecule. The helix-turn-helix motif with a recognition helix is present in the repressor, and moreover the repressor DNA-binding domains dimerize in the crystals in such a way that the recognition helices are separated by 34 A as in Cro. 

The requirements on building materials due to air velocities inside the building are generally negligible. However, sometimes the allowed contaminant concentrations can be of such magnitude that moving air may affect surfaces. In such cases it is necessary to use materials with sustainable surfaces. Normally this demand is valid only for the transport of dust-laden air in... 

Example An airflow model that has been validated for temperature differences of 15 K cannot be expected to predict smoke movement accurately in a building that is on fire with temperature differences that are 10 times larger. 

For a new process plant, calculations can be carried out using the heat release and plume flow rate equations outlined in Table 13.16 from a paper by Bender. For the theory to he valid, the hood must he more than two source diameters (or widths for line sources) above the source, and the temperature difference must be less than 110 °C. Experimental results have also been obtained for the case of hood plume eccentricity. These results account for cross drafts which occur within most industrial buildings. The physical and chemical characteristics of the fume and the fume loadings are obtained from published or available data of similar installations or established through laboratory or pilot-plant scale tests. - If exhaust volume requirements must he established accurately, small scale modeling can he used to augment and calibrate the analytical approach. 

One way to examine the validity of the steady-state approximation is to compare concentration—time curves calculated with exact solutions and with steady-state solutions. Figure 3-10 shows such a comparison for Scheme XIV and the parameters, ki = 0.01 s , k i = 1 s , 2 = 2 s . The period during which the concentration of the intermediate builds up from its initial value of zero to the quasi-steady-state when dcfjdt is vei small is called the pre-steady-state or transient stage in Fig. 3-10 this lasts for about 2 s. For the remainder of the reaction (over 500 s) the steady-state and exact solutions are in excellent agreement. Because the concen-... 

Our search procedures represent a departure from the above type of paradigm. Rather than simply accepting and implementing a decision policy found by DUg, that optimizes an overall measure of performance, the infimal subsystems and corresponding plant personnel play an active role in the construction and validation of solutions. One tries to build a consensus decision policy, Xpp, validated by all subsystems, DU , k = I,..., K, as well as by the whole plant, DUg, and only when that consensus has been reached does one move toward implementation. Within this context, the upper-level decision unit, DUg, assumes a eoordination role. 

---