Implementing two dimensional segment inversions with inversion-conforming data sets processing being rendered to include generalized composite weight factors in the processing of error-affected multivariate data samples

• US 7,383,128 B2
• Filed: 11/04/2005
• Issued: 06/03/2008
• Est. Priority Date: 11/12/2004
• Status: Expired due to Fees

• Current Assignee: Larry S. Chandler
• Original Assignee: Larry S. Chandler
• Inventors: Chandler, Larry S.
• Primary Examiner(s): Barlow; John
• Assistant Examiner(s): Taylor; Victor J.
• Application Number:
  US11/266,224
• Publication Number:
  US 20060111844A1
• Time in Patent Office: 942 Days
• Field of Search: 702 1- 18, 702/25
• US Class Current: 702/1
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04

Abstract

Representations of data inversions are generated by alternate forms of maximum likelihood estimating and associated least-squares and regression analysis which are rendered in correspondence with either single component residual deviations or projections between data samples and inversion-conforming data sets. Deficiencies in representing likelihood as related to errors-in-variables data and heterogeneous precision are compensated by composite weighting of likelihood elements. Composite weight factors employ both normalization to establish non-skewed homogeneous likelihood elements and fundamental weighting to compensate for associated non-linearly and establish common units for combining orthogonal coordinate-oriented data-point projections. Respective weight factors are related to alternately considered fundamental variables. Variance or alternate representation, as related to statistically independent sampling, is utilized as assumed applicable or replaced by composite variability representing single coordinate variations as affected by orthogonal coordinate sampling dispersions. Statistical rendition is generated as a replacement for unquantifiable dependent variable representation.

Claims

1. A method of using a data processing system for generating a data inversion in correspondence with at least one two dimensional segment comprising two dimensional data-related coordinates represented by a plurality of single-coordinate data samples and respective dependent variable correspondence;

said plurality of single-coordinate data samples being included in a set of coordinate designators comprising characteristic observation measurements selected from an ensemble of variable related observations;

said characteristic observation measurements corresponding to more independent variable coordinates than dependent variable coordinates,

said set of coordinate designators comprising at least one subset defined by the criteria that said subset excludes said plurality of single-coordinate data samples and said dependent variable correspondence, and comprises only respective coordinate related observation measurements that correspond to substantially constant variable measurements along respective variable coordinates and are presumed to be orthogonal to respective measurements corresponding to said plurality of single-coordinate data samples;

said substantially constant measurements being associated with said respective variable coordinates and represented by respective coordinate related observation measurements,

said respective coordinate related observation measurements being substantially constant as considered within prescribed limits, and

said data inversion comprising a representation of evaluated adjustment parameters;

said method comprising;

activating means for accessing processing and representing information, accessing provided data,

designating said at least one two dimensional segment, said designating including identifying said set of coordinate designators in correspondence with the criteria which define said subset;

abstracting said plurality of single-coordinate data samples from said ensemble of variable related observations in correspondence with said at least one two dimensional segment,

establishing a parametric approximative form in correspondence with said plurality of single-coordinate data samples and said dependent variable correspondence, and

effecting at least one form of data manipulation whereby said adjustment parameters are evaluated.

2. A method as claimed in claim 1, wherein said abstracting includes sequencing a representation of multivariate data samples to establish a plurality of two dimensional segments, said sequencing including:

rendering each of a plurality of said observation measurements by coordinate corresponding digits of respective integer representations, wherein each of said digits of each said respective integer corresponds within said prescribed limits and over a respectively considered variable range to said respectively considered variable sample coordinates, and

sequencing the respective integer representations of said observation measurements and isolating non-constant single coordinate data samples which correspond to substantially constant measurements associated with respective orthogonal variable coordinates.

3. A method as claimed in claim 1, wherein said dependent variable correspondence is represented by sample measurements included in said set of coordinate designators and excluded from said subset.

4. A method as claimed in claim 1, wherein said data manipulation includes rendering said at least one two dimensional segment in the absence of dependent variable sample measurements, and wherein said dependent variable correspondence is generated in correspondence with a sequenced digital representation of said plurality of single-coordinate data samples.

5. A method as claimed in claim 4,

wherein the ensemble of variable related observations represents data samples corresponding to only one dimension,

wherein said subset is a null subset, and

wherein said data inversion is rendered with respect to two dimensions;

one of said two dimensions corresponding to said plurality of single-coordinate data samples, and the other of said two dimensions corresponding to said dependent variable correspondence which is rendered in correspondence with said plurality of single-coordinate data samples.

6. A method as claimed in claim 1, wherein said data manipulation includes implementing proportionate composite weighting and generating said data inversion in correspondence with said implementing, said implementing including establishing said composite weighting in proportion to the respective products of fundamental weight factors multiplied times the squares of corresponding deviation normalization coefficients,

said deviation normalization coefficients rendering the products of un-normalized deviations multiplied by said deviation normalization coefficients so as to be substantially characterized by non-skewed homogeneous uncertainty distributions,

said fundamental weight factors being established in correspondence with normalized function deviations comprising the products of said un-normalized deviations multiplied by the respective deviation normalization coefficients;

said deviation normalization coefficients being considered as constant during the rendition of representation for the respective fundamental weight factors;

said data manipulation including optimizing said parametric approximative form in correspondence with a sum of addends, and

said addends being established as represented by the squares of said un-normalized deviations being rendered to include said proportionate composite weighting,

wherein successive approximations for said proportionate composite weighting are held constant during said optimizing.

7. A method as claimed in claim 6, wherein said un-normalized deviations are data-point projections and wherein said data manipulation includes effectuating at least one form of inversion-conforming data sets processing.

8. A method of using a data processing system for generating a data inversion in correspondence with at least one set of coordinate related measurements and a plurality of composite weight factors,

said at least one set of coordinate related measurements comprising a plurality of single-coordinate data samples,

said composite weight factors comprising products of fundamental weight factors multiplied times the squares of respective deviation normalization coefficients,

said deviation normalization coefficients rendering the products of un-normalized deviations multiplied by said deviation normalization coefficients so as to be substantially characterized by non-skewed homogeneous uncertainty distributions,

said fundamental weight factors being established in correspondence with normalized function deviations,

said normalized function deviations comprising products of said deviation normalization coefficients and the respective un-normalized deviations,

said deviation normalization coefficients being evaluated in correspondence with currently considered estimates for adjustment parameters,

said fundamental weight factors being related to products of change in said normalized function deviations,

said change in normalized function deviations being considered with respect to change in pertinent fundamental variables, and

said deviation normalization coefficients being considered as constant during the rendition of said change;

said method comprising;

establishing a parametric approximative form in correspondence with said plurality of single-coordinate data samples and respective dependent variable correspondence;

arranging a digitized representation of the normalized function deviations in a form for computer processing;

arranging a digitized representation of the fundamental weight factors in correspondence with said normalized function deviations in a form for computer processing;

determining said composite weight factors in proportion to the respective products of multiplying the fundamental weight factors times the squares of the corresponding deviation normalization coefficients;

establishing initial estimates for said adjustment parameters;

activating means for accessing processing and representing information;

accessing provided data;

evaluating initial estimates for the respective composite weight factors in correspondence with said plurality of single-coordinate data samples, said initial estimates for adjustment parameters, and said parametric approximative form, and

effecting at least one form of data manipulation whereby said adjustment parameters are evaluated in correspondence with a plurality of said un-normalized deviations and successive approximations for said composite weight factors, said effecting including representing information whereby at least one form of automated data processing is effectuated in correspondence with said parametric approximative form, and

said automated data processing including implementing at least one form of calculus of variation to optimize representation for said adjustment parameters in correspondence with a sum of addends,

said composite weight factors being considered as constant during the optimizing steps related to application of said at least one form of calculus of variation;

said addends being established as represented by the squares of said un-normalized deviations being rendered to include proportionate successive estimates for said composite weight factors, and

said data inversion comprising optimized representation for said adjustment parameters;

said at least one form of data manipulation excluding;

the generating and implementing of the square of inverse deviation variation weighting to establish the weighting of squared single component residual deviations,

the generating and implementing of inverse deviation variation weighting to establish the weighting of single component residual deviations,

the generating and implementing of cross term minimizing weight factors to establish the weighting of squared single component residual deviations,

the generating and implementing of transformation weight factors to establish the weighting of squared single component residual deviations, and

the representing and implementing of precision weighting as rendered in correspondence with forms of discriminate reduction data processing for the weighting of the squares of single component residual deviations.

9. A method as claimed in claim 8, wherein said un-normalized deviations are data-point projections and wherein said automated data processing includes rendering at least one form of inversion-conforming data sets processing;

said data inversion being generated in correspondence with a plurality of said data-point projections;

each of said data-point projections extending from respective data-related coordinates to intersect an approximating relationship;

intersections of said data-point projections with said approximating relationship substantially establishing respective inversion-conforming data sets which comprise projected coordinates for points that conform to a corresponding data inversion at the respective intersections.

10. A method as claimed in claim 8, wherein said data inversion is generated in correspondence with at least one two dimensional segment;

said at least one set of coordinate related measurements comprising an ensemble of variable related observations;

said at least one two dimensional segment comprising two dimensional data-related coordinates represented by said plurality of single-coordinate data samples and the respective dependent variable correspondence;

said plurality of single-coordinate data samples being included in a set of coordinate designators comprising characteristic observation measurements selected from said ensemble of variable related observations;

said characteristic observation measurements corresponding to more independent variable coordinates than dependent variable coordinates;

said set of coordinate designators comprising at least one subset defined by the criteria that said subset excludes said plurality of single-coordinate data samples and said dependent variable correspondence, and comprises only respective coordinate related observation measurements that correspond to substantially constant variable measurements along respective variable coordinates and are presumed to be orthogonal to respective measurements corresponding to said plurality of single-coordinate data samples;

said substantially constant measurements being associated with said respective variable coordinates and represented by the respective coordinate related observation measurements, and

the respective coordinate related observation measurements being established to be substantially constant as considered within prescribed limits;

said method comprising;

designating said at least one two dimensional segment, said designating including identifying said set of coordinate designators in correspondence with the criteria defining subset;

abstracting said plurality of single-coordinate data samples from said ensemble of variable related observations in correspondence with said at least one two dimensional segment, and

establishing parametric approximative form in correspondence with said plurality of single-coordinate data samples and said dependent variable correspondence.

11. A product comprising a memory device having stored thereon an electronic or electromagnetic representation of a data inversion generated by a data processing system,

said data inversion comprising a proportionate representation of an un-quantified dependent variable response associated with a plurality of single-coordinate data samples abstracted from an ensemble of variable related observations, said data inversion being rendered in correspondence with at least one two dimensional segment,

said two dimensional segment comprising two dimensional data-related coordinates represented by said plurality of single-coordinate data samples and respective dependent variable correspondence,

said plurality of single-coordinate data samples being included in a set of coordinate designators,

said set of coordinate designators comprising characteristic observation measurements selected from said ensemble of variable related observations,

said set of coordinate designators comprising at least one subset defined by the criteria that said subset excludes said plurality of single-coordinate data samples and said dependent variable correspondence and comprises only coordinate related observation measurements which are presumed to be orthogonal to respective measurements corresponding to said plurality of single-coordinate data samples and coordinate designators which correspond to substantially constant measurements along respective variable coordinates;

said substantially constant measurements being associated with said respective variable coordinates and represented by corresponding coordinate related observation measurements which are substantially constant within prescribed limits, and

said data inversion comprising a representation of evaluated adjustment parameters wherein said evaluated adjustment parameters are generated by;

activating means for accessing, processing and representing information,

accessing provided data,

representing information whereby at least one form of data processing is effectuated in correspondence with a parametric approximative form, and

effectuating said at least one form of data processing, said effectuating including;

designating said at least one two dimensional segment, said designating including identifying said set of coordinate designators in correspondence with the criteria defining said subset,

abstracting said plurality of single-coordinate data samples from said ensemble of variable related observations in correspondence with said at least one two dimensional segment,

establishing said parametric approximative form in correspondence with said plurality of single-coordinate data samples and said dependent variable correspondence, and

effecting at least one form of data manipulation whereby said adjustment parameters are evaluated, said effecting including processing said at least one two dimensional segment in the absence of dependent variable sample measurements;

wherein said dependent variable correspondence is generated in correspondence with a sequenced digital representation of said plurality of single-coordinate data samples.

12. A product as claimed in claim 11, wherein said abstracting includes sequencing a representation of multivariate data samples to establish a plurality of two dimensional segments, said sequencing including:

rendering each of a plurality of said observation measurements by coordinate corresponding digits of respective integer representations, wherein each of said digits of each said respective integer corresponds within said prescribed limits and over a respectively considered variable range to said respectively considered variable sample coordinates, and

sequencing the respective integer representations of said observation measurements and isolating non-constant single coordinate data samples which correspond to substantially constant measurements associated with respective orthogonal variable coordinates.

13. A product as claimed in claim 11, wherein said ensemble of variable related observations represents data samples corresponding to only one dimension,

wherein said subset is a null subset, and

wherein said data inversion is rendered with respect to two dimensions;

one of said two dimensions corresponding to said plurality of single-coordinate data samples, and the other of said two dimensions corresponding to said dependent variable correspondence which is rendered in correspondence with said plurality of single-coordinate data samples, and

said data inversion being rendered with disregard to proportionate offset values corresponding to said un-quantified dependent variable response.

14. A product as claimed in claim 11, wherein said effectuating includes implementing proportionate composite weighting,

said data inversion being generated in correspondence with said implementing, and

said implementing including establishing said composite weighting in proportion to the respective products of fundamental weight factors being multiplied times the squares of corresponding deviation normalization coefficients;

said deviation normalization coefficients rendering the products of un-normalized deviations multiplied by said deviation normalization coefficients so as to be substantially characterized by non-skewed homogeneous uncertainty distributions;

said fundamental weight factors being established in correspondence with normalized function deviations comprising the products of said un-normalized deviations multiplied by said deviation normalization coefficients;

said deviation normalization coefficients being considered as constant during the rendition of representation for the respective fundamental weight factors;

said data manipulation including optimizing said parametric approximative form in correspondence with a sum of addends;

said addends being established as represented by the squares of said un-normalized deviations being rendered to include said proportionate composite weighting,

wherein successive approximations for said proportionate composite weighting are held constant during said optimizing.

15. A data processing system comprising means for accessing, processing, and representing information whereby a data representation is generated in correspondence with at least one two dimensional segment comprising two dimensional data-related coordinates represented by a plurality of single-coordinate data samples and respective dependent variable correspondence;

said plurality of single-coordinate data samples being included in a set of coordinate designators comprising characteristic observation measurements selected from an ensemble of variable related observations;

said characteristic observation measurements corresponding to more independent variable coordinates than dependent variable coordinates,

said set of coordinate designators comprising at least one subset defined by the criteria that said subset excludes said plurality of single-coordinate data samples and said dependent variable correspondence, and comprises only respective coordinate related observation measurements that correspond to substantially constant variable measurements along respective variable coordinates and are presumed to be orthogonal to respective measurements corresponding to said plurality of single-coordinate data samples, and

said substantially constant measurements being associated with said respective variable coordinates and represented by the respective coordinate related observation measurements,

the respective coordinate related observation measurements being established to be substantially constant as considered within prescribed limits, and

said data representation comprising representation of evaluated adjustment parameters;

wherein said data representation is generated by;

activating the means for said accessing processing and representing information, accessing provided data,

designating said at least one two dimensional segment, said designating including identifying said set of coordinate designators in correspondence with criteria related to said subset;

abstracting said plurality of single-coordinate data samples from said ensemble of variable related observations in correspondence with said at least one two dimensional segment,

establishing parametric approximative form in correspondence with said plurality of single-coordinate data samples and said dependent variable correspondence, and

effecting at least one form of data manipulation whereby said adjustment parameters are evaluated.

16. A data processing system as claimed in claim 15, wherein said abstracting includes sequencing representations of multivariate data samples to establish a plurality of two dimensional segments, said sequencing including:

rendering each of a plurality of said observation measurements by coordinate corresponding digits of respective integer representations wherein each of said digits of each said respective integer corresponds within said prescribed limits and over a respectively considered variable range to said respectively considered variable sample coordinates, and

sequencing the respective integer representations of said observation measurements thereby isolating non-constant said single-coordinate data samples which correspond to substantially constant measurements being associated with respective orthogonal variable coordinates;

wherein said prescribed limits are determined by the variable range of the data samples, the characteristic base of the respective integer representations, and the selective designation of the respective two dimensional segments.

17. A data processing system as claimed in claim 15, wherein said included in said set of coordinate designators and excluded from said subset.

18. A data processing system as claimed in claim 15, wherein said data manipulation includes processing said at least one two dimensional segment in the absence of dependent variable sample measurements;

said dependent variable correspondence being generated in correspondence with a sequenced digital representation of said plurality of single-coordinate data samples.

19. A data processing system as claimed in claim 18,

wherein the ensemble of variable related observations represents data samples corresponding to only one dimension;

wherein said subset is a null subset;

wherein said data representation is rendered with respect to two dimensions corresponding to said plurality of single-coordinate data samples, and

wherein said dependent variable correspondence is rendered in correspondence with said plurality of single-coordinate data samples.

20. A data processing system as claimed in claim 15, wherein

said data manipulation includes implementing proportionate composite weighting;

said data representation is generated in correspondence with said implementing;

said implementing includes establishing said composite weighting in proportion to the respective multiplication products of fundamental weight factors times the square of corresponding deviation normalization coefficients;

said deviation normalization coefficients rendering the products of un-normalized deviations multiplied by said deviation normalization coefficients so as to be substantially characterized by non-skewed homogeneous uncertainty distributions;

said fundamental weight factors being established in correspondence with normalized function deviations comprising the products of said un-normalized deviations multiplied by the respective deviation normalization coefficients;

said deviation normalization coefficients being considered as constant during the rendition of representation for the respective fundamental weight factors;

said data manipulation including optimizing said parametric approximative form in correspondence with a sum of addends, and

said addends being established as represented by the square of said un-normalized deviations being rendered to include said proportionate composite weighting;

wherein successive approximations for said proportionate composite weighting are held constant during said optimizing.

21. A data processing system as claimed in claim 16 comprising a composite weight factor generator comprising means for rendering composite weighting in correspondence with the multiplication product of the fundamental weight factors times the square of corresponding deviation normalization coefficients;

said fundamental weight factors being proportional to a root of the inverse of the product of derivatives of normalized function deviations taken with respect to pertinent system variables,

said root corresponding to the number of said pertinent system variables, and

said deviation normalization coefficients being considered as constant during the rendition of said derivatives.