| LTPDA Toolbox™ | contents | ![]() |
SPSDSUBTRACTION makes a sPSD-weighted least-square iterative fit %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% DESCRIPTION: SPSDSUBTRACTION makes a sPSD-weighted least-square iterative fit CALL: [MPest, aoResiduum, plOut, aoP, aoPini] = optSubtraction(mat_Y, mat_U); [MPest, aoResiduum, plOut, aoP, aoPini] = optSubtraction(mat_Y, mat_U, pl); The function finds the optimal M that minimizes the sum of the weighted sPSD of (mat_Y - M * mat_U) OUTPUTS: - MPest: output PEST object with parameter estimates - aoResiduum: residuum times series - plOut: plist containing data like the parameter estimates - aoP: last weight used in the optimization (fater last Maximization/Expectation step) - aoPini: initial weight used in the optimization (before first Maximization/Expectation step) Parameters Description %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
| Method Details | |
|---|---|
| Access | public |
| Defining Class | matrix |
| Sealed | 0 |
| Static | 0 |
| Sets for this method … |
|---|
| Default |
Default |
|||
|---|---|---|---|
| no description | |||
| Key | Default Value | Options | Description |
| spsdSubtraction | |||
| COEFS | [] | none | initial subtracted coefficients, must be a nY*nU double array. If not provided zeros are assumed |
| WEIGHTINGMETHOD | 'residual' |
|
choose to define a frequency weighting scheme |
| AOWEIGHT | [] | none | ao to define a frequency weighting scheme (if chosen in 'weightingMethod') |
| PZMODELWEIGHT | [] | none | pzmodel to define a frequency weighting scheme (if chosen in 'weightingMethod') |
| LINCOEF | 5 | none | linear coefficient for scaling frequencies in chi2 |
| LOGCOEF | 0.29999999999999999 | none | logarithmic coefficient for scaling frequencies in chi2 |
| ITERMAX | 20 | none | max number of Mex/Exp iterations |
| NORMCOEFS | 1.0000000000000001e-15 | none | tolerance on inf norm of coefficient update (used depending on 'CVCriterion') |
| NORMCRITERION | 1.0000000000000001e-15 | none | tolerance on norm of criterion variation (used depending on 'CVCriterion') |
| WIN | (WIN='levelledHanning', PSLL=200, LEVELORDER=4) | none | window to operate FFT, may be a plist/ao |
Example |
|---|
| plist('COEFS', [[]], 'WEIGHTINGMETHOD', 'residual', 'AOWEIGHT', [[]], 'PZMODELWEIGHT', [[]], 'LINCOEF', [5], 'LOGCOEF', [0.29999999999999999], 'ITERMAX', [20], 'NORMCOEFS', [1.0000000000000001e-15], 'NORMCRITERION', [1.0000000000000001e-15], 'WIN', plist('WIN', 'levelledHanning', 'PSLL', [200], 'LEVELORDER', [4])) |
back to top
| Some information of the method matrix/spsdSubtraction are listed below: | |
|---|---|
| Class name | matrix |
| Method name | spsdSubtraction |
| Category | Operator |
| Package name | ltpda |
| Can be used as modifier | 1 |
| Supported numeric types | {'double'} |
|
Method: matrix/simplify | Method: matrix/char | ![]() |
©LTP Team