Specialized data analysis

There are two ways in which data can be analyzed using the MARS system. The first is through use of the general data analysis tools which are part of the standard MARS shell. The second is by means of specialized data analysis programming which can be invoked by the MARS shell. 

For specialized data analysis such as DMA mechanical properties or DSC reaction kinetics, only the bare bones analysis code need be written and Interfaced to MARS. 

In addition, several special data analysis tools are available through Tools => Data Analysis. While most of these are for statistical and business use, we will use two of them, for Random Number Generation, and for Regression. Data Analysis also contains a Fourier Analysis tool, which we will not use because a simpler macro is provided, see chapters 7 and 9. Likewise, the Regression tool can be replaced by the Weighted Least Squares macro discussed in chapters 3 and 9, which is somewhat simpler to use but does not provide as much statistical information. 

These results demonstrate the synergism that is possible in a detailed, multitechnique approach to a problem where no one technique (XPS, AES, or SEM) would have given a complete understanding. They also illustrate the usefulness of specialized data analysis, such as that provided by SBDs. 

NPRDS data are available to users, either through annual summary reports (to resume publication in the coming year) or through direct on-line data base access from a computer terminal. Special reports and listings are available through specific requests for extraction or data analysis. 

Opening segments of the IP2 PRA data analysis section describe the definitions of terms and concepts employed, the assumptions made, and limitations recognized during the data base construction. A set of 39 plant-specific component failure mode summaries established the basis for component service hour determinations, the number of failures, and the test data source for each failure mode given for each component. Generic data from WASH-1400, IEEE Std 500, and the LER data summaries on valves, pumps, and diesels were combined with plant-specific failure data to produce "updated" failure information. All the IP2 specialized component hardware failure data, both generic and updated, are contained in Table 1.5.1-4 (IP3 1.6.1-4). This table contains (by system, component, and failure mode) plant-specific data on the number of failures and service hours or demands. For some components, it was determined that specifications of the system was warranted because of its impact on the data values. 

A theorem, which we do not prove here, states that the nonzero eigenvalues of the product AB are identical to those of BA, where A is an nxp and where B is a pxn matrix [3]. This applies in particular to the eigenvalues of matrices of cross-products XX and X which are of special interest in data analysis as they are related to dispersion matrices such as variance-covariance and correlation matrices. If X is an nxp matrix of rank r, then the product X X has r positive eigenvalues in A and possesses r eigenvectors in V since we have shown above that ... 

Single value charts are only used for special purposes, e.g. as original value chard for the determination of warning and control limits or, for data analysis of time series (Shumway [1988] Montgomery et al. [1990]). All the other types of charts are used relatively often and have their special advantages (Besterfield [1979] Montgomery [1985] Wheeler and Chambers [1990]). 

Table I is a list of physical properties of materials which were of special concern, along with target values felt to indicate useful levels in a particular application. From the beginning it was predicted that one of the biggest problems would be to balance Properties A and E, usually considered mutually exclusive. It was also assumed that Properties B and E were highly correlated. Statistically designed experiments and data analysis were chosen to determine most efficiently the formulations which would give the best combination of all the target properties.

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. 

The distance between object points is considered as an inverse similarity of the objects. This similarity depends on the variables used and on the distance measure applied. The distances between the objects can be collected in a distance matrk. Most used is the euclidean distance, which is the commonly used distance, extended to more than two or three dimensions. Other distance measures (city block distance, correlation coefficient) can be applied of special importance is the mahalanobis distance which considers the spatial distribution of the object points (the correlation between the variables). Based on the Mahalanobis distance, multivariate outliers can be identified. The Mahalanobis distance is based on the covariance matrix of X this matrix plays a central role in multivariate data analysis and should be estimated by appropriate methods—mostly robust methods are adequate. 

Regarding point 2, the information extraction function of chemometrics is a very valnable one that is often overlooked, especially in the industrial world. It will be mentioned later in this chapter that this function can be nsed concurrently with the instrument specialization function, rather than relying npon additional exploratory data analysis. 

The second area of progressive development will be in the area of very specialized data collection mediums. We foresee a major extension in the area of direct data input using interactive video terminals built within the sensory evaluation booths. Although this will take a fairly extensive development in the area of software, the gains to be reaped by the development will far exceed the development costs itself. A large area of research has yet to be conducted concerning the application of microcomputers in sensory analysis as a data collection tool. Studies of inherent bias, effects on response freedom etc. will have to be undertaken in order to properly evaluate the computer as a data acquisition tool. Biases based on the way the software... 

Furthermore it is recommended that NMR newcomers start with the central volume Processing Strategies and complete their education in modern NMR spectroscopy according to their special needs by working through the appropriate volumes, Data Acquisition, Modern Data Analysis and Intelligent Data Management. 

Today s analytical end users are made up of (1) researchers who specialize in analysis and (2) researchers from other pharmaceutical specialties. Researchers who specialize in analysis (i.e., analytical chemists) with formal training in the fundamentals of instrumentation, method development, and the treatment of data are continually challenged with relearning the fundamentals of chemistry (Laitinen, 1980) and other pharmaceutical sciences, such as molecular biology. They must keep current with advances in the field as well as the pharmaceutical business in general. 

---