Method mfh/loglikelihood
LOGLIKELIHOOD: Compute log-likelihood for MFH objects
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Compute the (unnormalized) log-likelihood for MFH objects.
For the MFH objects, the log-likelihood function must be defined
by the user. For example, for the definition <o - H x i>*S^-1<o - H x i>
the function 'func' must cover the term <o - H x i>. The 'S' matrix
corresponds to the 'noise' plist key.
EXAMPLE: LLH = loglikelihood(func, pl);
For more than one channel model, then input an array of MFH objects.
NOTE: The 'S' matrix must be of the correct size Nout X Nout.
For example, for the 2 input - 2 output case:
mfh_fncs = [ch1_mfh_exp1 , ch1_mfh_exp2 ; ...
ch2_mfh_exp1 , ch2_mfh_exp2 ]
LLH = loglikelihood(mfh_fncs, plist);
For systems that require multiple channels and experiments, then:
LLH = loglikelihood([ch1_mfh ; ch2_mfh], plist);
OUTPUTS: logL - A collection of objects containing the LLH, SNR,
LLH(frequencies).
NOTE: If callerIsMethod is true, only the numerical values of the
above items are returned.
Example: [LLH SNR LLH(f)] = loglikelihood(m, plist);
Otherwise, if callerIsMethod == false
collection = loglikelihood(myfunc, plist);
Parameters Description
M. Nofrarias, N. Karnesis 2012
| Method Details |
|
| Access |
public |
| Defining Class |
mfh |
| Sealed |
0 |
| Static |
0 |
Parameter Description
| Sets for this method … |
| Default |
Default |
| no description |
| Key |
Default Value |
Options |
Description |
| loglikelihood |
| X |
[] |
none |
The parameter values. A 1xNumParams array. |
| NOISE |
[] |
none |
The inverse cross-spectrum matrix of the measured noise. |
| F |
[] |
none |
Numerical array of frequencies. |
| WRAP DOUBLE |
0 |
|
Wraps the input model in a double() call to ensure the model output is numeric for internal calculations. If your model already returns a numeric vector, you can leave this set to false. |
| K0 |
1 |
none |
The first FFT coefficient of the analysis. All FFT coefficients with k
|
Example |
| plist('X', [[]], 'NOISE', [[]], 'F', [[]], 'WRAP DOUBLE', [false], 'K0', [1]) |
back to top
| Some information of the method mfh/loglikelihood are listed below: |
| Class name |
mfh |
| Method name |
loglikelihood |
| Category |
Signal Processing |
| Package name |
ltpda |
| Can be used as modifier |
1 |
| Supported numeric types |
{'double'} |
|
Method: mfh/lincom |
|
Method: mfh/multinest |
 |
©LTP Team