 Introducing Auguri™Published Date: 2008-06-21 23:05:37 WorkOnInternet.com
Data Inspection, Editing and Visualization Auguri allows opening multiple document files simultaneously for editing, inspection, charting, analysis or forecasting. It maps data to worksheets, where a worksheet may consist of one or several sheets containing additional data, models, solutions and reports. Data may be edited freely and charted at will in most popular formats. Not only can you instruct Auguri to plot your data in 3-dimensions, but produce fully functional four-dimensional representations that can be zoomed, rotated, or even animated, if you so desire. Auguri also includes modern interpolation techniques for the replacement of missing values, or for resampling of existing data where additional information is desirable before continuing with further analysis. In addition, Auguri provides several methods for the automatic generation of constant and serial data in your documents. Statistical, Frequency, Linear and Nonlinear Analysis Other than common statistical analysis methods such as ANOVA, test of means and variances, histograms, IID tests and descriptive statistics to the fourth moment, Auguri provides extensive nonlinear methods such as generalized fractal dimensions, Poincare surface of sections, maximal Lyapunov exponent, false nearest neighbors, space-time separation plots, averaged mutual information, and phase portraits in up to four-dimensions, among other. Auguri also provides tools for the analysis of signals and series in the time and frequency domains, such as power spectrum estimation, Fourier transforms, auto- and cross- covariance and correlation functions, time-evolving statistics, and simultaneous solutions to linear equations. In addition, Auguri includes several methods for the generation of random numbers according to a chosen distribution, for sampling data from existing populations, and for generating surrogate data, where statistical and nonlinearity tests may be additionally carried. Data Preparation, Transformation and Signal Separation Apart from standard operations such as scaling, normalization, differencing, summation and operations between series, Auguri includes a powerful functional expression parser. Simply type your functions following a left to right order of evaluation as you would write them in paper, and save them. You can then apply these functions directly to selected ranges of data through all of your open documents for as long as you want. You can also design custom frequency filters that can be applied to data in the time or frequency domain, separate your signals at the chosen frequencies, and allow for the specification of distinct models that can provide an enhancement in the accuracy of your forecasts and systems approximations. Auguri also includes methods for the empirical differentiation and integration of data for instances when these quantities are preferable or desirable in a model. Smoothing and Noise Reduction Auguri lists several methods for data smoothing and noise reduction. Some of these methods are explicit, as in the case of moving averages, frequency convolution, exponential smoothing and Savitzky-Golay. Other methods will require the definition of an intermediate model where an auto-regressive predictor, a locally constant predictor, or an artificial neural network may be used as the discriminant function. These methods provide a good set of choices for reducing noise in instances where the preservation of the nonlinear dynamics of chaotic attractors in state space is desirable. Model Specification With Auguri, system models may be specified with a few clicks of the mouse. There is no need to expand data to specify complex, past and multivariate dependencies. All data expansion takes place automatically in memory, following your specifications, at prediction or approximation time. You can also save models and methods from any of your Auguri documents, load them to other documents, or simply have them stored as one of the possible solutions to a system for future references. Solutions and Comparisons Auguri allows the specification of concurrent solutions to defined models, which can be tested in a simulation, compared, or individually ran with new input data. Select global or local methods guided by fast and precise state of the art spatial databases, and variations thereof. Define solutions according to your preferences and data characteristics. Compare auto-regressive with local least squares methods side-by-side, or against powerful multilayer feedforward neural networks trained with innovative quasi-unsupervised learning algorithms. Any and all is a choice. No restrictions apply. Reports and Data Export It is always good to know exactly how calculations take place. For this reason Auguri reports the quantities of each and every operation it carries. If not explicitly, there is always a tool that will allow you to request so. Neural interconnection weights, approximation errors, least squares coefficients, model arguments—all available. You can copy, save or print charts, format and print documents. It's all your choice. Finally, you can export your files to most common ANSI and binary formats, or drag and drop (or copy and paste) content to your favorite editor or spreadsheet program. Auguri is a trademark of Advanced Analytics Group LLC Windows is a trademark of Microsoft Corporation
Home | Submit Article| Article Topics | Article Archives | Newsletter | Business Directory | Contact US |
