Abstract: A computer system performs data processing using a data organizer which parses a primary set of data into predefined sets of data components according to one or more enterprise models. The sets of data components from the data organizer are respectively stored in multiple data sources. The data components may be stored as a binary file or hierarchical file. A plurality of data processing systems execute the enterprise models. A task manager controls which enterprise model executes on each data processing system. The data processing systems access the data components from the data sources which correspond the executing enterprise model. The data components are sent over dedicated communication channels to the data processing systems. Each data processing system may have a data interface, which is configured to receive data from different sources and convert the data into a standard format.

Abstract: A computer-based profit optimization model is provided that takes account of supply-side and demand-side factors in optimizing profit for an organization. The model takes into account which parts the organization uses to assemble various products. Demand curves are used to characterize the quantity of each product that will be demanded for various classes of product and classes of customer. Supply model data is used to determine which mix of products can be sold in view of parts availability. Using the demand model and supply model data, the optimization model can recommend a set of prices to use for selling the organization's products to various customers and at various shipment times. The model ensures that the organization has sufficient resources available to produce the products and enforces user-supplied business rules and other constraints.

Abstract: A profit optimization system takes account of supply-side and demand-side factors in optimizing profit for an organization. The profit optimization system uses an optimization model to optimize profit in a spot market. The model takes into account which parts the organization uses to assemble various products. Demand curves are used to characterize the quantity of each product that will be demanded as a function of price on the spot market. Supply model data is used to determine which mix of products can be sold in view of parts availability. Using the demand model and supply model data, the optimization model can recommend a set of prices to use for selling the organization's products. The model ensures that the organization has sufficient resources available to produce the products and enforces user-supplied business rules and other constraints.

Abstract: A computer system models customer response using observable data. The observable data includes transaction, product, price, and promotion. The computer system receives data observable from customer responses. A set of factors including customer traffic within a store, selecting a product, and quantity of selected product is defined as expected values, each in terms of a set of parameters related to customer buying decision. A likelihood function is defined for each of the set of factors. The parameters are solved using the observable data and associated likelihood function. The customer response model is time series of unit sales defined by a product combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product. A linear relationship is given between different products which includes a constant of proportionality that determines affinity and cannibalization relationships between the products.

Abstract: A bid evaluation tool is provided for evaluating proposed bids for commodity sales contracts. The tool displays historical price data and forecasts in a graph. A user of the tool defines a date range for a proposed contract by entering starting and ending dates for the contract. The tool analyzes price data for the date range and generates a corresponding proposed bid for the date range. A bid window may be used to gather user inputs for the tool. The user can select between multiple bid pricing models for the bid evaluation tool to use when processing the price data to produce the proposed bid. Bid models for the tool include models such as a flat average model, a linear fit model, and an offset model. The user can adjust which historical data is included in the graph and can select which data set is used to generate bids.

Abstract: A computer system models customer response using observable data. The observable data includes transaction, product, price, and promotion. The computer system receives data observable from customer responses. A set of factors including customer traffic within a store, selecting a product, and quantity of selected product is defined as expected values, each in terms of a set of parameters related to customer buying decision. A likelihood function is defined for each of the set of factors. The parameters are solved using the observable data and associated likelihood function. The customer response model is time series of unit sales defined by a product combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product. A linear relationship is given between different products which includes a constant of proportionality that determines affinity and cannibalization relationships between the products.

Abstract: A model creation tool is provided that automatically evaluates potential econometric models and, given a set of historical data on which the model is intended to operate and other constraints, automatically identifies an optimum model. The optimum model may be an autoregressive model. The optimum model is defined by a set of optimum model parameters and optimum model coefficients. The optimum model parameters may include a model order such as an autoregressive model order, a seasonality function parameter, a trend function parameter, and a parameter that sets a tolerance that is used when performing principal components dimension reduction operations on the historical data.

Abstract: A computer system performs data processing using a data organizer which parses a primary set of data into predefined sets of data components according to one or more enterprise models. The sets of data components from the data organizer are respectively stored in multiple data sources. The data components may be stored as a binary file or hierarchical file. A plurality of data processing systems execute the enterprise models. A task manager controls which enterprise model executes on each data processing system. The data processing systems access the data components from the data sources which correspond the executing enterprise model. The data components are sent over dedicated communication channels to the data processing systems. Each data processing system may have a data interface, which is configured to receive data from different sources and convert the data into a standard format.

Abstract: A computer system performs data processing using a data organizer which parses a primary set of data into predefined sets of data components according to one or more enterprise models. The sets of data components from the data organizer are respectively stored in multiple data sources. The data components may be stored as a binary file or hierarchical file. A plurality of data processing systems execute the enterprise models. A task manager controls which enterprise model executes on each data processing system. The data processing systems access the data components from the data sources which correspond the executing enterprise model. The data components are sent over dedicated communication channels to the data processing systems. Each data processing system may have a data interface, which is configured to receive data from different sources and convert the data into a standard format.

Abstract: A complex signal correlator such as can be implemented in a correlation detector system affords a unique algorithm which estimates the codifference correlation between two complex signals based on the sum and difference of codifference estimates, each codifference estimate having equivalently associated therewith a dispersion estimate. Typical embodiments provide a receiving antenna and a receiver inclusive of the codifference correlator, wherein radio frequency waves are down converted and sampled, the sampled signals are correlated with a reference signal contained in a memory, and the resultant correlation signal is detected and transduced. The inventive correlator is based on an alpha-stable distribution and, in comparison with conventional alpha-stable distribution-based correlators, can more effectively operate in a realm wherein alpha is less than one.

Abstract: A signal detection system divides a data sampling run into blocks and perms a fast Fourier transform on each block, sorting results by frequency. Combinations of results of the fast Fourier transform corresponding to each frequency are processed to derive a test statistic which is unbiased by Gaussian noise while including such combinations of results of the fast Fourier transform which would be redundant over other combinations. Information concerning the frequency behavior of the signal derived in the course of detection, is accomplished with increased sensitivity.

Abstract: A signal detection system, preferably for use with a coherent radar system elects certain combination of input signal samples in a block of signal samples to derive a test statistic which is unbiased by Gaussian noise. Products of pairs of sample values are stored in a data sample look-up table. An index look-up table is created by scanning through possible combinations of addresses and excluding those combinations of sample values which would be redundant over other combination and would result in a contribution to biasing by noise. Pairs of addresses from the index look-up table are then used in sequence to access first and second products from the data sample look-up table. The first product is multiplied by the complex conjugate of the second product and the (quadruple product) result is averaged to form a test statistic which is insensitive to position and constant velocity. Detection of the signal is determined by testing the real part of the averaged test statistic against a threshold.

Abstract: While the computation of high resolution radar images based on higher order tatistics is position insensitive, velocity estimation may be based on the ratios of values of a trispectral slice and a cross-trispectral slice computed as quadruple products of complex valued signals developed by coherent radar. Signal-to-noise ratio is improved by either averaging over a plurality of bursts during computation of both the trispectral slice and the cross-trispectral slice or averaging of values of ratios of trispectral slice and cross-trispectral slice values at particular frequencies or wavenumbers, or both.

Abstract: A coherent radar is used for achieving high resolution radar imaging of a moving object in sea clutter and noise. A stored replica of the transmitted waveform is combined with the returned signal in a synchronous detector to produce in-phase and quadrature (I and Q or complex) samples. High resolution is achieved by transmitting a series of pulses, each at a different frequency, and then processing these complex samples to produce a synthetic down-range profile. To enhance the radar target or object from system noise and sea clutter, the complex I and Q (in-phase and quadrature component) samples from the radar are coherently averaged in a special two-dimensional slice of the trispectrum (e.g. quadruple product or fourth-order moment). This formulation of the fourth moment retains all important target information and suppresses Gaussian noise. Once averaged, a new set of I and Q samples are reconstructed that produce the same trispectral slice as this average.