atmopy.stat.measure

Package atmopy :: Package stat :: Module measure

[show private | hide private]

Module atmopy.stat.measure

Function Summary
float	`bf(data1, data2, cutoff)` Computes Bias Factor (BF) of data1 and data2.
float	`correlation(data1, data2)` Computes the correlation between data1 and data2.
float	`determination(data1, data2)` Computes the coefficient of determination between data1 and data2.
float	`fac2(data1, data2)` Computes the proportion of values of data1 within a factor 2 of the values of data2.
float	`fac5(data1, data2)` Computes the proportion of values of data1 within a factor 5 of the values of data2.
float	`fge(data1, data2, cutoff)` Computes Fractional Gross Error (FE) between data1 and data2 1D arrays.
float	`fmt(data1, data2)` Computes the figure of merit in time of data2 and data1.
float	`mage(data1, data2)` Computes Mean Absolute Gross Error (MAGE) between data1 and data2.
float	`mbe(data1, data2)` Computes Mean Bias Error between data1 and data2.
float	`mfbe(data1, data2, cutoff)` Computes Mean Fractionalized Bias Error (MFBE) between data1 and data2 1D arrays.
float	`mg(data1, data2, cutoff)` Computes Geometric Mean Bias (mg) between data1 and data2.
float	`mnbe(data1, data2, cutoff)` Computes Mean Normalized Bias Error (MNBE) between data1 and data2 1D arrays.
float	`mnge(data1, data2, cutoff)` Computes Mean Normalized Gross Error (MNGE) between data1 and data2 1D arrays.
float	`nmb(data1, data2)` Computes Normalized Mean Bias (NMB) between data1 and data2.
float	`nme(data1, data2)` Computes Normalized Mean Error (NME) between data1 and data2.
float	`nmse_1(data1, data2)` Computes Normalized Mean Square Error (NMSE_1) between data1 and data2.
float	`rmse(data1, data2)` Computes Root Mean Square Error (RMSE) between data1 and data2.
float	`rnmse_2(data1, data2, cutoff)` Computes the Square Root of the second version of the Normalized Mean Square Error (RNMSE_2) between data1 and data2.
float	`upa(data1, data2)` Computes Unpaired Peak Accuracy between data1 and data2.
float	`vg(data1, data2, cutoff)` Computes Geometric Variance (vg) between data1 and data2.

Function Details

bf(data1, data2, cutoff=0.0)

Computes Bias Factor (BF) of data1 and data2.

Parameters:: data1 - 1D array to compute BF.
(type=numpy.array); data2 - 1D array to compute BF.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and corresponding 'data1' values.
(type=float)

Returns:: Bias Factor of data1 and data2.
(type=float)

correlation(data1, data2)

Computes the correlation between data1 and data2.

Parameters:: data1 - 1D array to compute correlation.
(type=numpy.array); data2 - 1D array to compute correlation.
(type=numpy.array)

Returns:: Correlation coefficient between data1 and data2.
(type=float)

determination(data1, data2)

Computes the coefficient of determination between data1 and data2. This is the correlation coefficient squared.

Parameters:: data1 - 1D array to compute coefficient of determination.
(type=numpy.array); data2 - 1D array to compute coefficient of determination.
(type=numpy.array)

Returns:: Coefficient of determination between data1 and data2.
(type=float)

fac2(data1, data2)

Computes the proportion of values of data1 within a factor 2 of the values of data2.

Parameters:: data1 - 1D array to compute fac2.
(type=numarray.array); data2 - 1D array to compute fac2.
(type=numarray.array)

Returns:: The proportion of data1 values within a factor 2 of data2 values.
(type=float)

fac5(data1, data2)

Computes the proportion of values of data1 within a factor 5 of the values of data2.

Parameters:: data1 - 1D array to compute fac5.
(type=numarray.array); data2 - 1D array to compute fac5.
(type=numarray.array)

Returns:: The proportion of data1 values within a factor 5 of data2 values.
(type=float)

fge(data1, data2, cutoff=0.0)

Computes Fractional Gross Error (FE) between data1 and data2 1D arrays.

Parameters:: data1 - 1D array to compute FE.
(type=numpy.array); data2 - 1D array to compute FE.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and corresponding 'data1' values.
(type=float)

Returns:: Fractional Gross Error between data1 and data2.
(type=float)

fmt(data1, data2)

Computes the figure of merit in time of data2 and data1.

Parameters:: data1 - 1D array to compute fb.
(type=numpy.array); data2 - 1D array to compute fb.
(type=numpy.array)

Returns:: The figure of merit in time of data2 and data1.
(type=float)

mage(data1, data2)

Computes Mean Absolute Gross Error (MAGE) between data1 and data2.

Parameters:: data1 - 1D array to compute error.
(type=numpy.array); data2 - 1D array to compute error.
(type=numpy.array)

Returns:: Mean Absolute Gross Error between data1 and data2.
(type=float)

mbe(data1, data2)

Computes Mean Bias Error between data1 and data2.

Parameters:: data1 - 1D array to compute bias from.
(type=numpy.array); data2 - 1D array to compute bias from.
(type=numpy.array)

Returns:: Mean Bias Error between data1 and data2.
(type=float)

mfbe(data1, data2, cutoff=0.0)

Computes Mean Fractionalized Bias Error (MFBE) between data1 and data2 1D arrays.

Parameters:: data1 - 1D array to compute MFBE.
(type=numpy.array); data2 - 1D array to compute MFBE.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and corresponding 'data1' values.
(type=float)

Returns:: Mean Fractionalized Bias Error between data1 and data2.
(type=float)

mg(data1, data2, cutoff=0.0)

Computes Geometric Mean Bias (mg) between data1 and data2.

Parameters:: data1 - 1D array to compute NMSE.
(type=numpy.array); data2 - 1D array to compute NMSE.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and 'data1' values.
(type=float)

Returns:: Geometric Mean Bias between data1 and data2.
(type=float)

mnbe(data1, data2, cutoff=0.0)

Computes Mean Normalized Bias Error (MNBE) between data1 and data2 1D arrays.

Parameters:: data1 - 1D array to compute MNBE.
(type=numpy.array); data2 - 1D array to compute MNBE.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and corresponding 'data1' values.
(type=float)

Returns:: Mean Normalized Bias Error between data1 and data2.
(type=float)

mnge(data1, data2, cutoff=0.0)

Computes Mean Normalized Gross Error (MNGE) between data1 and data2 1D arrays. ( MNGE = 1 / n * NME )

Parameters:: data1 - 1D array to compute error.
(type=numpy.array); data2 - 1D array to compute error.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and corresponding 'data1' values.
(type=float)

Returns:: Mean Normalized Gross Error (MNGE) between data1 and data2.
(type=float)

nmb(data1, data2)

Computes Normalized Mean Bias (NMB) between data1 and data2.

Parameters:: data1 - 1D array to compute NMB.
(type=numpy.array); data2 - 1D array to compute NMB.
(type=numpy.array)

Returns:: Normalized Mean Bias between data1 and data2.
(type=float)

nme(data1, data2)

Computes Normalized Mean Error (NME) between data1 and data2.

Parameters:: data1 - 1D array to compute NME.
(type=numpy.array); data2 - 1D array to compute NME.
(type=numpy.array)

Returns:: Normalized Mean Error between data1 and data2.
(type=float)

nmse_1(data1, data2)

Computes Normalized Mean Square Error (NMSE_1) between data1 and data2.

Parameters:: data1 - 1D array to compute NMSE_1.
(type=numarray.array); data2 - 1D array to compute NMSE_1.
(type=numarray.array)

Returns:: First version of Normalized Mean Square Error between data1 and data2.
(type=float)

rmse(data1, data2)

Computes Root Mean Square Error (RMSE) between data1 and data2.

Parameters:: data1 - 1D array to compute error.
(type=numpy.array); data2 - 1D array to compute error.
(type=numpy.array)

Returns:: Root Mean Square Error (RMSE) between data1 and data2.
(type=float)

rnmse_2(data1, data2, cutoff=0.0)

Computes the Square Root of the second version of the Normalized Mean Square Error (RNMSE_2) between data1 and data2.

Parameters:: data1 - 1D array to compute RNMSE_2.
(type=numpy.array); data2 - 1D array to compute RNMSE_2.
(type=numpy.array)

Returns:: Square Root of the second version of the Normalized Mean Square Error between data1 and data2.
(type=float)

upa(data1, data2)

Computes Unpaired Peak Accuracy between data1 and data2. This can be paired or unpaired peak prediction accuracy depending on simulated data used (interpolated or not..)

Parameters:: data1 - 1D array to compute Peak Estimation Accuracy.
(type=numpy.array); data2 - 1D array to compute Peak Estimation Accuracy.
(type=numpy.array)

Returns:: Unpaired Peak Accuracy of data1 and data2.
(type=float)

vg(data1, data2, cutoff=0.0)

Computes Geometric Variance (vg) between data1 and data2.

Parameters:: data1 - 1D array to compute NMSE.
(type=numpy.array); data2 - 1D array to compute NMSE.
(type=numpy.array); cutoff - The value below (or equal) which data is discarded. This filters 'data2' and corresponding 'data1' values.
(type=float)

Returns:: Geometric Variance between data1 and data2.
(type=float)

Generated by Epydoc 2.1 on Tue May 19 10:37:12 2009

http://epydoc.sf.net