Parameter estimation multiple measurements

The most reliable estimates of the parameters are obtained from multiple measurements, usually a series of vapor-liquid equilibrium data (T, P, x and y). Because the number of data points exceeds the number of parameters to be estimated, the equilibrium equations are not exactly satisfied for all experimental measurements. Exact agreement between the model and experiment is not achieved due to random and systematic errors in the data and due to inadequacies of the model. The optimum parameters should, therefore, be found by satisfaction of some selected statistical criterion, as discussed in Chapter 6. However, regardless of statistical sophistication, there is no substitute for reliable experimental data. 

The true model parameters (ft, ftj(...) are partial derivatives of the response function / and cannot be measured directly. It is, however, possible to otain estimates, ft, bV], bVl, of these parameters by multiple regression methods in which the polynomial model is fitted to known experimental results obtained by varying the settings of xr. These variations will then define an experimental design and are conveniently displayed as a design matrix, D, in which the rows describe the settings in the individual experiments and the columns describe the variations of the experimental variables over the series of experiments. 

The parameter estimation presented in the previous section is based on a least squares minimisation of the errors between measured system outputs and outputs of a system model evaluated by using estimated parameter values. If the real system is replaced by a model in preferred integral causality, measured outputs can be obtained by solving the model equations for given initial conditions and can be used for offline parameter estimation in order to isolate multiple faults deliberately introduced into the system model. In real-time FDI, initial conditions are either not known or difficult to obtain. Therefore, in online parameter estimation, they have to be considered as additional unknowns that are to be estimated. 

A cost function to be minimised in an iterative parameter estimation procedure may be formulated by using either differences between outputs from a real system and computed outputs from a model or by means of ARR residuals. As output errors, as well as ARR residuals are generally nonlinear functions of the component parameters, multiple fault parameter isolation becomes a well-known nonlinear least squares problem. For real-time FDI, ARR residuals obtained from a DBG have the advantage that they make the parameter estimation independent of any initial conditions of the process that are hardly known and will have to be estimated along with component parameters. In off-line simulation, the real system may be replaced by a behavioural model. Measured data is then generated by assuming realistic consistent initial conditions and by solving the equations of the behavioural model. 

At the present time 28 research groups are participating in this study. It is planned to double the number of the groups in order to treat 8,000 patients, estimated to afford statistically significant evaluation of the results. The primary end-point is survival. Multiple parameters will be measured in the patients during a 5 year period. 

When significant intakes may have occurred, more refined calculations based on individual specific parameters (special dosimetry) should be made (Section 3). If multiple measurements are available, a single best estimate of intake may be obtained, for example, by the method of least squares [41, 42]. 

Many biochemists use the velocity equations for kinetic parameter estimates despite the fact that the rates are difficult to determine experimentally. In practice either the substrate depletion or the product formation is measured as a function of time and the rates are calculated by differentiating the data, leading to an inexact analysis (Schnell Mendoza, 1997,2000a). Alternatively, the differential equations governing the biochemical reactions may be solved or approximated to obtain reactant concentration as function of time. This approach decreases the number of experimental assays by at least a factor of live, as proved by Schnell and Mendoza (2001), because multiple experimental points may be collected for each single reaction. 

The quantities AUMC and AUSC can be regarded as the first and second statistical moments of the plasma concentration curve. These two moments have an equivalent in descriptive statistics, where they define the mean and variance, respectively, in the case of a stochastic distribution of frequencies (Section 3.2). From the above considerations it appears that the statistical moment method strongly depends on numerical integration of the plasma concentration curve Cp(r) and its product with t and (r-MRT). Multiplication by t and (r-MRT) tends to amplify the errors in the plasma concentration Cp(r) at larger values of t. As a consequence, the estimation of the statistical moments critically depends on the precision of the measurement process that is used in the determination of the plasma concentration values. This contrasts with compartmental analysis, where the parameters of the model are estimated by means of least squares regression. 

Because PB-PK models are based on physiological and anatomical measurements and all mammals are inherently similar, they provide a rational basis for relating data obtained from animals to humans. Estimates of predicted disposition patterns for test substances in humans may be obtained by adjusting biochemical parameters in models validated for animals adjustments are based on experimental results of animal and human in vitro tests and by substituting appropriate human tissue sizes and blood flows. Development of these models requires special software capable of simultaneously solving multiple (often very complex) differential equations, some of which were mentioned in this chapter. Several detailed descriptions of data analysis have been reported. 

To verify such a steric effect a quantitative structure-property relationship study (QSPR) on a series of distinct solute-selector pairs, namely various DNB-amino acid/quinine carbamate CSPpairs with different carbamate residues (Rso) and distinct amino acid residues (Rsa), has been set up [59], To provide a quantitative measure of the effect of the steric bulkiness on the separation factors within this solute-selector series, a-values were correlated by multiple linear and nonlinear regression analysis with the Taft s steric parameter Es that represents a quantitative estimation of the steric bulkiness of a substituent (Note s,sa indicates the independent variable describing the bulkiness of the amino acid residue and i s.so that of the carbamate residue). For example, the steric bulkiness increases in the order methyl < ethyl < n-propyl < n-butyl < i-propyl < cyclohexyl < -butyl < iec.-butyl < t-butyl < 1-adamantyl < phenyl < trityl and simultaneously, the s drops from -1.24 to -6.03. In other words, the smaller the Es, the more bulky is the substituent. The obtained QSPR equation reads as follows ... 

Decision levels and detection limits are relatively easy to define and evaluate for simple" (zero dimensional) measurements. The transition to higher dimensions and multiple components introduces a number of complications and added assumptions related to the number and identity of components, shapes and parameters of calibration functions and spectra, and distributional consequences of non-linear estimation. 

Except for very simple systems, initial rate experiments of enzyme-catalyzed reactions are typically run in which the initial velocity is measured at a number of substrate concentrations while keeping all of the other components of the reaction mixture constant. The set of experiments is run again a number of times (typically, at least five) in which the concentration of one of those other components of the reaction mixture has been changed. When the initial rate data is plotted in a linear format (for example, in a double-reciprocal plot, 1/v vx. 1/[S]), a series of lines are obtained, each associated with a different concentration of the other component (for example, another substrate in a multisubstrate reaction, one of the products, an inhibitor or other effector, etc.). The slopes of each of these lines are replotted as a function of the concentration of the other component (e.g., slope vx. [other substrate] in a multisubstrate reaction slope vx. 1/[inhibitor] in an inhibition study etc.). Similar replots may be made with the vertical intercepts of the primary plots. The new slopes, vertical intercepts, and horizontal intercepts of these replots can provide estimates of the kinetic parameters for the system under study. In addition, linearity (or lack of) is a good check on whether the experimental protocols have valid steady-state conditions. Nonlinearity in replot data can often indicate cooperative events, slow binding steps, multiple binding, etc. 

Natural budworm densities were determined by sampling 6 sprays, each 40 cm long, In the same quarter of the tree used to collect tissue for chemical analysis and to collect defoliation data. Densities were expressed as the average number of budworm larvae per 100 buds per tree. A visual estimate of the amount of defoliation eilso was made In the same area of the crown where the densities and needle tissue were collected. Since budworm may disperse from heavily defoliated trees, (Greenback, 1963) budworm densities from each tree were weighted by the level of defoliation that each tree sustained. This resulted In an Infestation Intensity measurement (dependent variable) which was subjected to multiple stepwise correlation analysis using various foliage quality and physical tree parameters as the Independent variables. Thirty-one parameters were used as Independent variables In this analysis. 

---