--- /dev/null
+# Created by Octave 3.6.1, Wed Mar 28 20:33:11 2012 UTC <root@brouzouf>
+# name: cache
+# type: cell
+# rows: 3
+# columns: 33
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 15
+average_moments
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 35
+ for internal use by gmm_estimate
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 35
+ for internal use by gmm_estimate
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+delta_method
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 92
+ Computes Delta method mean and covariance of a nonlinear
+ transformation defined by "func"
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ Computes Delta method mean and covariance of a nonlinear
+ transformation define
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+gmm_estimate
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 927
+ usage: [theta, obj_value, convergence, iters] =
+ gmm_estimate(theta, data, weight, moments, momentargs, control, nslaves)
+
+ inputs:
+ theta: column vector initial parameters
+ data: data matrix
+ weight: the GMM weight matrix
+ moments: name of function computes the moments
+ (should return nXg matrix of contributions)
+ momentargs: (cell) additional inputs needed to compute moments.
+ May be empty ("")
+ control: (optional) BFGS or SA controls (see bfgsmin and samin).
+ May be empty ("").
+ nslaves: (optional) number of slaves if executed in parallel
+ (requires MPITB)
+
+ outputs:
+ theta: GMM estimate of parameters
+ obj_value: the value of the gmm obj. function
+ convergence: return code from bfgsmin
+ (1 means success, see bfgsmin for details)
+ iters: number of BFGS iteration used
+
+ please type "gmm_example" while in octave to see an example
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage: [theta, obj_value, convergence, iters] =
+ gmm_estimate(theta,
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+gmm_example
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 126
+ GMM example file, shows initial consistent estimator,
+ estimation of efficient weight, and second round
+ efficient estimator
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ GMM example file, shows initial consistent estimator,
+ estimation of efficient
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 7
+gmm_obj
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 206
+ The GMM objective function, for internal use by gmm_estimate
+ This is scaled so that it converges to a finite number.
+ To get the chi-square specification
+ test you need to multiply by n (the sample size)
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ The GMM objective function, for internal use by gmm_estimate
+ This is scaled so
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+gmm_results
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 1145
+ usage: [theta, V, obj_value] =
+ gmm_results(theta, data, weight, moments, momentargs, names, title, unscale, control, nslaves)
+
+ inputs:
+ theta: column vector initial parameters
+ data: data matrix
+ weight: the GMM weight matrix
+ moments: name of function computes the moments
+ (should return nXg matrix of contributions)
+ momentargs: (cell) additional inputs needed to compute moments.
+ May be empty ("")
+ names: vector of parameter names
+ e.g., names = char("param1", "param2");
+ title: string, describes model estimated
+ unscale: (optional) cell that holds means and std. dev. of data
+ (see scale_data)
+ control: (optional) BFGS or SA controls (see bfgsmin and samin). May be empty ("").
+ nslaves: (optional) number of slaves if executed in parallel
+ (requires MPITB)
+
+ outputs:
+ theta: GMM estimated parameters
+ V: estimate of covariance of parameters. Assumes the weight matrix
+ is optimal (inverse of covariance of moments)
+ obj_value: the value of the GMM objective function
+
+ please type "gmm_example" while in octave to see an example
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage: [theta, V, obj_value] =
+ gmm_results(theta, data, weight, moments, mome
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+gmm_variance
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 49
+ GMM variance, which assumes weights are optimal
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 49
+ GMM variance, which assumes weights are optimal
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 24
+gmm_variance_inefficient
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 53
+ GMM variance, which assumes weights are not optimal
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 53
+ GMM variance, which assumes weights are not optimal
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 14
+kernel_density
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 1178
+ kernel_density: multivariate kernel density estimator
+
+ usage:
+ dens = kernel_density(eval_points, data, bandwidth)
+
+ inputs:
+ eval_points: PxK matrix of points at which to calculate the density
+ data: NxK matrix of data points
+ bandwidth: positive scalar, the smoothing parameter. The fit
+ is more smooth as the bandwidth increases.
+ kernel (optional): string. Name of the kernel function. Default is
+ Gaussian kernel.
+ prewhiten bool (optional): default false. If true, rotate data
+ using Choleski decomposition of inverse of covariance,
+ to approximate independence after the transformation, which
+ makes a product kernel a reasonable choice.
+ do_cv: bool (optional). default false. If true, calculate leave-1-out
+ density for cross validation
+ computenodes: int (optional, default 0).
+ Number of compute nodes for parallel evaluation
+ debug: bool (optional, default false). show results on compute nodes if doing
+ a parallel run
+ outputs:
+ dens: Px1 vector: the fitted density value at each of the P evaluation points.
+
+ References:
+ Wand, M.P. and Jones, M.C. (1995), 'Kernel smoothing'.
+ http://www.xplore-stat.de/ebooks/scripts/spm/html/spmhtmlframe73.html
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 55
+ kernel_density: multivariate kernel density estimator
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 22
+kernel_density_cvscore
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 38
+ some kernels can assign zero density
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 38
+ some kernels can assign zero density
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 20
+kernel_density_nodes
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 87
+ kernel_density_nodes: for internal use by kernel_density - does calculations on nodes
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ kernel_density_nodes: for internal use by kernel_density - does calculations on
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 14
+kernel_example
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 161
+ kernel_example: examples of how to use kernel density and regression functions
+ requires the optim and plot packages from Octave Forge
+
+ usage: kernel_example;
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ kernel_example: examples of how to use kernel density and regression functions
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 24
+kernel_optimal_bandwidth
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 309
+ kernel_optimal_bandwidth: find optimal bandwith doing leave-one-out cross validation
+ inputs:
+ * data: data matrix
+ * depvar: column vector or empty ("").
+ If empty, do kernel density, orherwise, kernel regression
+ * kernel (optional, string) the kernel function to use
+ output:
+ * h: the optimal bandwidth
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ kernel_optimal_bandwidth: find optimal bandwith doing leave-one-out cross valid
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 17
+kernel_regression
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 1100
+ kernel_regression: kernel regression estimator
+
+ usage:
+ fit = kernel_regression(eval_points, depvar, condvars, bandwidth)
+
+ inputs:
+ eval_points: PxK matrix of points at which to calculate the density
+ depvar: Nx1 vector of observations of the dependent variable
+ condvars: NxK matrix of data points
+ bandwidth (optional): positive scalar, the smoothing parameter.
+ Default is N ^ (-1/(4+K))
+ kernel (optional): string. Name of the kernel function. Default is
+ Gaussian kernel.
+ prewhiten bool (optional): default true. If true, rotate data
+ using Choleski decomposition of inverse of covariance,
+ to approximate independence after the transformation, which
+ makes a product kernel a reasonable choice.
+ do_cv: bool (optional). default false. If true, calculate leave-1-out
+ fit to calculate the cross validation score
+ computenodes: int (optional, default 0).
+ Number of compute nodes for parallel evaluation
+ debug: bool (optional, default false). show results on compute nodes if doing
+ a parallel run
+ outputs:
+ fit: Px1 vector: the fitted value at each of the P evaluation points.
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 48
+ kernel_regression: kernel regression estimator
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 23
+kernel_regression_nodes
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 93
+ kernel_regression_nodes: for internal use by kernel_regression - does calculations on nodes
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ kernel_regression_nodes: for internal use by kernel_regression - does calculati
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+mle_estimate
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 758
+ usage:
+ [theta, obj_value, conv, iters] = mle_estimate(theta, data, model, modelargs, control, nslaves)
+
+ inputs:
+ theta: column vector of model parameters
+ data: data matrix
+ model: name of function that computes log-likelihood
+ modelargs: (cell) additional inputs needed by model. May be empty ("")
+ control: (optional) BFGS or SA controls (see bfgsmin and samin). May be empty ("").
+ nslaves: (optional) number of slaves if executed in parallel (requires MPITB)
+
+ outputs:
+ theta: ML estimated value of parameters
+ obj_value: the value of the log likelihood function at ML estimate
+ conv: return code from bfgsmin (1 means success, see bfgsmin for details)
+ iters: number of BFGS iteration used
+
+ please see mle_example.m for examples of how to use this
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage:
+ [theta, obj_value, conv, iters] = mle_estimate(theta, data, model, mode
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+mle_example
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 42
+ Example to show how to use MLE functions
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 42
+ Example to show how to use MLE functions
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 7
+mle_obj
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 178
+ usage: [obj_value, score] = mle_obj(theta, data, model, modelargs, nslaves)
+
+ Returns the average log-likelihood for a specified model
+ This is for internal use by mle_estimate
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 77
+ usage: [obj_value, score] = mle_obj(theta, data, model, modelargs, nslaves)
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 13
+mle_obj_nodes
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+ Who am I?
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+ Who am I?
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+mle_results
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 918
+ usage: [theta, V, obj_value, infocrit] =
+ mle_results(theta, data, model, modelargs, names, title, unscale, control)
+
+ inputs:
+ theta: column vector of model parameters
+ data: data matrix
+ model: name of function that computes log-likelihood
+ modelargs: (cell) additional inputs needed by model. May be empty ("")
+ names: vector of parameter names, e.g., use names = char("param1", "param2");
+ title: string, describes model estimated
+ unscale: (optional) cell that holds means and std. dev. of data (see scale_data)
+ control: (optional) BFGS or SA controls (see bfgsmin and samin). May be empty ("").
+ nslaves: (optional) number of slaves if executed in parallel (requires MPITB)
+
+ outputs:
+ theta: ML estimated value of parameters
+ obj_value: the value of the log likelihood function at ML estimate
+ conv: return code from bfgsmin (1 means success, see bfgsmin for details)
+ iters: number of BFGS iteration used
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage: [theta, V, obj_value, infocrit] =
+ mle_results(theta, data, model, mo
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+mle_variance
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 122
+ usage: [V,scorecontribs,J_inv] =
+ mle_variance(theta, data, model, modelargs)
+
+ This is for internal use by mle_results
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage: [V,scorecontribs,J_inv] =
+ mle_variance(theta, data, model, modelargs)
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+nls_estimate
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 780
+ usage:
+ [theta, obj_value, conv, iters] = nls_estimate(theta, data, model, modelargs, control, nslaves)
+
+ inputs:
+ theta: column vector of model parameters
+ data: data matrix
+ model: name of function that computes the vector of sums of squared errors
+ modelargs: (cell) additional inputs needed by model. May be empty ("")
+ control: (optional) BFGS or SA controls (see bfgsmin and samin). May be empty ("").
+ nslaves: (optional) number of slaves if executed in parallel (requires MPITB)
+
+ outputs:
+ theta: NLS estimated value of parameters
+ obj_value: the value of the sum of squared errors at NLS estimate
+ conv: return code from bfgsmin (1 means success, see bfgsmin for details)
+ iters: number of BFGS iteration used
+
+ please see nls_example.m for examples of how to use this
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage:
+ [theta, obj_value, conv, iters] = nls_estimate(theta, data, model, mode
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+nls_example
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 56
+
+ define arguments for nls_estimate #
+
+ starting values
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 38
+
+ define arguments for nls_estimate #
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 7
+nls_obj
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 185
+ usage: [obj_value, score] = nls_obj(theta, data, model, modelargs, nslaves)
+
+ Returns the average sum of squared errors for a specified model
+ This is for internal use by nls_estimate
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 77
+ usage: [obj_value, score] = nls_obj(theta, data, model, modelargs, nslaves)
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 13
+nls_obj_nodes
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 42
+ This is for internal use by nls_estimate
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 42
+ This is for internal use by nls_estimate
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 12
+parameterize
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 316
+ usage: theta = parameterize(theta, otherargs)
+
+ This is an empty function, provided so that
+ delta_method will work as is. Replace it with
+ the parameter transformations your models use.
+ Note: you can let "otherargs" contain the model
+ name so that this function can do parameterizations
+ for a variety of models
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ usage: theta = parameterize(theta, otherargs)
+
+ This is an empty function, pro
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 7
+poisson
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 65
+ Example likelihood function (Poisson for count data) with score
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 65
+ Example likelihood function (Poisson for count data) with score
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 15
+poisson_moments
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 53
+ the form a user-written moment function should take
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 53
+ the form a user-written moment function should take
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 11
+prettyprint
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 49
+ this prints matrices with row and column labels
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 49
+ this prints matrices with row and column labels
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 13
+prettyprint_c
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 59
+ this prints matrices with column labels but no row labels
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 59
+ this prints matrices with column labels but no row labels
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 10
+scale_data
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 69
+ Standardizes and normalizes data matrix,
+ primarily for use by BFGS
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 69
+ Standardizes and normalizes data matrix,
+ primarily for use by BFGS
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 17
+sum_moments_nodes
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 34
+ for internal use by gmm_estimate
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 34
+ for internal use by gmm_estimate
+
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 18
+unscale_parameters
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 86
+ Unscales parameters that were estimated using scaled data
+ primarily for use by BFGS
+
+
+
+# name: <cell-element>
+# type: sq_string
+# elements: 1
+# length: 80
+ Unscales parameters that were estimated using scaled data
+ primarily for use by
+
+
+
+
+
git://git.creatis.insa-lyon.fr / CreaPhase.git / blobdiff