Regression analysis, initial estimates

The MO concentrations versus time profiles were fitted to second order polynomial equations and the parameters estimated by nonlinear regression analysis. The initial rates of reactions were obtained by taking the derivative at t=0. The reaction is first order with respect to hydrogen pressure changing to zero order dependence above about 3.45 MPa hydrogen pressure. This was attributed to saturation of the catalyst sites. Experiments were conducted in which HPLC grade MIBK was added to the initial reactant mixture, there was no evidence of product inhibition. 

One ean use the linearized least squares analysis to obtain initial estimates of the parameters k, Kj, K, in order to obtain convergence in nonlinear regression. However, in many eases it is possible to use a nonlinear regression analysis direetly as deseribed m Section 5.4 and in Example 5-6. 

Regression analysis is often employed to fit experimental data to a mathematical model. The purpose may be to determine physical properties or constants (e.g., rate constants, transport coefficients), to discriminate between proposed models, to interpolate or extrapolate data, etc. The model should provide estimates of the uncertainty in calculations from the resulting model and, if possible, make use of available error in the data. An initial model (or models) may be empirical, but with advanced knowledge of reactors, distillation columns, other separation devices, heat exchangers, etc., more sophisticated and fundamental models can be employed. As a starting point, a linear equation with a single independent variable may be initially chosen. Of importance, is the mathematical model linear In general, a function,/, of a set of adjustable parameters, 3y, is linear if a derivative of that function with respect to any adjustable parameter is not itself a function of any other adjustable parameter, that is. 

In nonlinear regression analysis, we search for those parameter t alues that minimize the sum of the squares of the differences beiw een the measured values and the calculated values for all the data points.- Not only can nonlinear regression find the best estimates of parameter values, it can al,so be used to discriminate between different rate law models, such as the Langmutr-Hin-shelw ood models discussed in Chapter 10. Many software programs are available to find these parameter values so that all one has to do is enter the data, The Polymath software will be used to illustrate this technique. In order to carry out the search efficiently, in some cases one has to enter initial estimates of the parameter -alues close to the actual values. These estimates can be obtained using Ihe linear-least-squares technique discussed on the CD-ROM Professional Reference Shelf. 

Supported is based on the average measurement of °Pb activity determined by measuring the activity of a °Pb as a decay product such as Pb in the lowest section of the sediment profile where °Pb activity is constant. Sedimentation or accretion rates are estimated from the excess (unsupported) °Pb profiles in the sediment profile using the constant initial concentration method (Goldberg et al., 1977). Linear regression analysis is used to solve for (X/s) in the log-transformed equation for radioactive decay ... 

Duggleby, R.G. (1985). Estimation of the initial velocity of enzyme-catalysed reactions by non-linear regression analysis of progress curves. The Bkchemkal Journal, vol. 228, no.l, (May 1985), pp. 55-60, ISSN 0264-6021... 

However, the regression theory requires that the errors be normally distributed around (—7 a). and not around f as in the linearized version just described. Hence use the values determined as initial estimates to obtain more accurate values of the constants by minimizing the sum of squares of the residuals of the rates directly from the raw rate equation by nonlinear least squares analysis. 

Another surprise was that interstrand cross-Unks were also formed independently of O2. However, this was initially rationalized by demonstrating that O2 reacted reversibly with 102 (Scheme 43). Nonlinear regression analysis of the ratio of thymidine to oxygenated products (eg 107) as a function of GSH concentration provided an estimated rate constant for GSH trapping of 102 (fecsH = 6.9 X 10 M s ) consistent with expectations for reaction of an alkyl radical with the thiol.The accuracy of the thiol trapping rate constant validated the extracted rate constant for O2 loss from 106 to reform 102 (feo2 = d.4 s ), which is consistent with rate constants reported for O2 loss from other peroxyl radicals. 

The gradient method applied to the objective function of a regression analysis with a model consisting of two parameters, / i and / 2- The initial estimates are represented by point A. From there, it goes zigzag to B, C, D, E, F, and further (indicated by dots). The method using Booth s modification evolves faster to the minimum from A to B, C, D, E, F, and further (indicated by filled squares). 

When the nature and composition of the sample is not well known, it is necessary to use influence correction methods, of which there are three primary types fundamental, derived, and regression. In the fundamental approach, the intensity of fluorescence can be calculated for each element in a standard sample from variables such as the source spectrum, the fundamental eiiuations for absorption and fluorescence, matrix effects the crystal reflectivity (in a WDXRF instrument), instrument aperture, the detector efficiency, and so forth. The XRF spectrum of the standard is measured, and in an iterative process the instrument variables are refined and combined with the fundamental variables to obtain a calibration function for the analysis. Then the spectrum for an unknown sample is measured, and the iterative process is repeated using initial estimates of the concentrations of the analytes. Iteration continues until the calculated spectrum matches the unknown spectrum according to appropriate statistical criteria. This method gives good results with accuracies on the order of I %-4% but is generally considered to be less accurate than derived... 

For the regression analysis, a reparameterized form of the Arrhenius and Van t Hoff equations was used. A full statistical analysis, which included the calculation of the 95% confidence intervals on the estimated parameters, was performed after regression. Initially, the response curves were recorded using a sampling time of 1 ms. For a correct statistical analysis the sampling time had to be increased to 8 ms. 

---