| 60 | | Allowing dense-sparse operations with CholeskiSolve |
| 61 | | New built-in function Sort |
| 62 | | New thinnig parameter for BinReadRows |
| 63 | | New built-in function MatReadDimensions |
| 64 | | New VMatrix functions DifEq and BackDifEq with or without specific initial values |
| 65 | | Allowing convertion between VMatrix Blas.R.Dense and Cholmod.R.Sparse and viceverse |
| 66 | | Fixed bug in default argument Y0 for DifEq and BackDifEq |
| 67 | | Fixed bug in PreProdDiag |
| 68 | | Applying faster matrix cycling methos to some functions |
| 69 | | Forzing LLt transformation and other MatLab lchol options |
| 70 | | Adding methods related to difference equations |
| 71 | | New methods IsFinite nd InUnknown |
| 72 | | New built-in functions Pol2VMat, SubBand |
| 73 | | Forzing Structure Alignement to 8 for cholmod.h |
| 74 | | Checking dimensions on MatProd |
| 75 | | New internal method replicateTriang |
| 76 | | Fixed bug in dpotri that was returned as triangular instead of symmetric |
| 77 | | Fixed bug inBDBClassify |
| 78 | | New TOL data type VMatrix to handle with sparse and dense matrices all together. This type is |
| | 60 | 1. Allowing dense-sparse operations with CholeskiSolve |
| | 61 | 1. New built-in function Sort |
| | 62 | 1. New thinnig parameter for BinReadRows |
| | 63 | 1. New built-in function MatReadDimensions |
| | 64 | 1. New VMatrix functions DifEq and BackDifEq with or without specific initial values |
| | 65 | 1. Allowing convertion between VMatrix Blas.R.Dense and Cholmod.R.Sparse and viceverse |
| | 66 | 1. Fixed bug in default argument Y0 for DifEq and BackDifEq |
| | 67 | 1. Fixed bug in PreProdDiag |
| | 68 | 1. Applying faster matrix cycling methos to some functions |
| | 69 | 1. Forzing LLt transformation and other MatLab lchol options |
| | 70 | 1. Adding methods related to difference equations |
| | 71 | 1. New methods IsFinite nd InUnknown |
| | 72 | 1. New built-in functions Pol2VMat, SubBand |
| | 73 | 1. Forzing Structure Alignement to 8 for cholmod.h |
| | 74 | 1. Checking dimensions on MatProd |
| | 75 | 1. New internal method replicateTriang |
| | 76 | 1. Fixed bug in dpotri that was returned as triangular instead of symmetric |
| | 77 | 1. Fixed bug inBDBClassify |
| | 78 | 1. New TOL data type VMatrix to handle with sparse and dense matrices all together. This type is |
| 80 | | * Current matrix subtypes are CHOLMOD sparse, factor and triplet and BLAS column major dense. |
| 81 | | * System has been designed to be easily upgraded with new subtypes as Toeplitz, Hankel, |
| 82 | | Vandermonde, ... |
| 83 | | * There are all kind of matrix related operations, not all developed already, that has been |
| 84 | | implemented to be the more efficient and the more transparent for the end user. System |
| 85 | | can convert internally, if needed, from dense to sparse, sparse to triplet, etc.; to make |
| 86 | | an operation be posible or faster between differently stored matrices |
| 87 | | - arithmetic: sum, rest, product, ... |
| 88 | | - linear algebra: Choleski decomposition and system solving, ... |
| 89 | | - access and building: cell, band and block access, concatenation, ... |
| 90 | | - input/output: (including OIS storement) etc) |
| 91 | | API of TOL for using BLAS and LAPACK functions |
| 92 | | *Using BLAS dgemm in generic matrix products |
| 93 | | *Using BLAS dsyrk in MtMSqr |
| 94 | | *Using BLAS dtrmm in triangular matrix products |
| 95 | | *Using BLAS dtrsm in TrSolve, LTSolve and LTInverse |
| 96 | | *Using LAPACK dpotrf in choleski method |
| 97 | | *Using LAPACK dpotrf and dpotri in CholeskiInverse |
| 98 | | *Using BLAS-LAPACK methods in LinearRegression and LinReg |
| 99 | | Using open source ATLAS as default implementation of BLAS and LAPACK |
| 100 | | Adding new method InnerPoint to get a point matching a system of linear inequations |
| 101 | | Fixing bug 522. Ensuring access to valid cells in Matrix Sub. |
| 102 | | Added warning message of invalid access at MatDat and PutMatDat |
| | 80 | 1. Current matrix subtypes are CHOLMOD sparse, factor and triplet and BLAS column major dense. |
| | 81 | 1. System has been designed to be easily upgraded with new subtypes as Toeplitz, Hankel, Vandermonde, ... |
| | 82 | 1. There are all kind of matrix related operations, not all developed already, that has been implemented to be the more efficient and the more transparent for the end user. System can convert internally, if needed, from dense to sparse, sparse to triplet, etc.; to make an operation be posible or faster between differently stored matrices |
| | 83 | 1. arithmetic: sum, rest, product, ... |
| | 84 | 1. linear algebra: Choleski decomposition and system solving, ... |
| | 85 | 1. access and building: cell, band and block access, concatenation, ... |
| | 86 | 1. input/output: (including OIS storement) etc) |
| | 87 | 1. API of TOL for using BLAS and LAPACK functions |
| | 88 | 1. Using BLAS dgemm in generic matrix products |
| | 89 | 1. Using BLAS dsyrk in MtMSqr |
| | 90 | 1. Using BLAS dtrmm in triangular matrix products |
| | 91 | 1. Using BLAS dtrsm in TrSolve, LTSolve and LTInverse |
| | 92 | 1. Using LAPACK dpotrf in choleski method |
| | 93 | 1. Using LAPACK dpotrf and dpotri in CholeskiInverse |
| | 94 | 1. Using BLAS-LAPACK methods in LinearRegression and LinReg |
| | 95 | 1. Using open source ATLAS as default implementation of BLAS and LAPACK |
| | 96 | 1. Adding new method InnerPoint to get a point matching a system of linear inequations |
| | 97 | 1. Fixing bug 522. Ensuring access to valid cells in Matrix Sub. |
| | 98 | 1. Added warning message of invalid access at MatDat and PutMatDat |
| 105 | | New Quantile API to get a set of quantiles saving time |
| 106 | | New built-in function ARIMAAlmagroEval for evaluation of logarithm of likelihood of ARIMA models |
| 107 | | Avoiding white noise model in ARMAAutoCovarianzeVector |
| 108 | | Using faster LevinsonARMA instead of standard Levinson |
| 109 | | Fixed bug calculating logLH in ARIMALevinsonEval. The error did not change the maximum of likelihood. |
| 110 | | -New StdLib::Rkde: API between TOL and R KDE (Kernel Density Estimation) |
| 111 | | -New StdLib::BysMcmc: Generic Bayesian MonteCarlo Markov Chain Methods and applications |
| 112 | | -New StdLib::BysMcmc:BSR (Bayesian Sparse Regression) |
| 113 | | *Estimates a sparse linear, constrained or not, regression by MCMC Gibbs method |
| 114 | | *Can use Boost Spirit to parse a human-like-formated description of the model |
| 115 | | *Stores data as sparse CHOLMOD matrix and implements all steps |
| 116 | | *The system is designed to be easily upgraded to handle with non linearities |
| | 101 | 1. New Matrix Quantile API to get a set of quantiles saving time |
| | 102 | 1. New built-in function ARIMAAlmagroEval for evaluation of logarithm of likelihood of ARIMA models |
| | 103 | 1. Avoiding white noise model in ARMAAutoCovarianzeVector |
| | 104 | 1. Using faster LevinsonARMA instead of standard Levinson |
| | 105 | 1. Fixed bug calculating logLH in ARIMALevinsonEval. The error did not change the maximum of likelihood. |
| | 106 | 1. ARIMA MLE method Estimate is internally evaluated in local scope to avoid export unexpected symbols |
| | 107 | 1. BLR:New strategy simulation that offers a much better efficiency without any loss of accuracy or numerical strength [https://www.tol-project.org/browser/tolp/trunk/tol/stdlib/math/stat/models/bayesian/linear/_blrc_kernel.tol 1. NameBlok's to call R |
| | 108 | 1. Rapi: TOL API for R [https://www.tol-project.org/browser/tolp/trunk/tol/stdlib/math/R stdlib/math/R] |
| | 109 | 1. Rglpk: an interface to the GNU Linear Programing Kit for solving large-scale linear programming (LP), mixed integer linear programming (MILP) and other related problems [https://www.tol-project.org/browser/tolp/trunk/tol/stdlib/math/optim/LP stdlib/math/optim/LP] |
| | 110 | 1. Rquadprog: Functions to solve Quadratic Programming Problems (stdlib/math/optim/QP) |
| | 111 | 1. RCoda: output analysis and diagnostics for MonteCarlo Markov Chains (MCMC) (stdlib/math/stat/models/bayesian/diagnosis) |
| | 112 | 1. New StdLib::Rkde: API between TOL and R KDE (Kernel Density Estimation) |
| | 113 | 1. New StdLib::BysMcmc: Generic Bayesian MonteCarlo Markov Chain Methods and applications |
| | 114 | 1. New StdLib::BysMcmc:BSR (Bayesian Sparse Regression) |
| | 115 | 1. Estimates a sparse linear, constrained or not, regression by MCMC Gibbs method |
| | 116 | 1. Can use Boost Spirit to parse a human-like-formated description of the model |
| | 117 | 1. Stores data as sparse CHOLMOD matrix and implements all steps |
| | 118 | 1. The system is designed to be easily upgraded to handle with non linearities |
| 118 | | BSR new blocks: Input and output missing, ARIMA |
| 119 | | Handling with vectorial or time series data in a transparent way |
| 120 | | Non linear filers of output and input |
| 121 | | Handling with normal prior information about parameters of non linear filters. Prior information is given as a log density additive function |
| 122 | | Avoiding to store all nodes to save memory |
| 123 | | Enhanced SQL scripting |
| 124 | | Storing failed factorization matrix |
| 125 | | Allowing fixed sigma in observational nodes |
| 126 | | User can specify the number of simulations to be resumed (do.resume>0) or las simulations to be skiped (do.resume<0) |
| 127 | | Adding SLICE scalar sampler |
| 128 | | New API OneNode to define simple BSR models with just one observational node and optional prior information and order relations |
| 129 | | New non linear filter for probit models |
| 130 | | Ensuring input matrix has changed. Else it will reuse old factorization |
| 131 | | Allowing Inverse-Gamma prior over sigma^2 of BSR segments |
| 132 | | Using faster method LogDens.Almagro.Z_cond_U in draw.ARMS |
| 133 | | New BSR specialization DynHlm for Dynamic Hierarchical Linear Model |
| 134 | | New general API for importing model definition |
| 135 | | New methods BysMcmc::Bsr::Gibbs::Estim and Bsr::Gibbs::Instance(...)::eval |
| 136 | | Adding optional time info assigned to each segment of noise |
| 137 | | Adding and grouping documental information |
| 138 | | New AlgLib Hypothesis testing: |
| 139 | | *AlgLib.JarqueBeraTest |
| 140 | | *AlgLib.StudentTtest1 |
| 141 | | *AlgLib.StudentTtest2 |
| 142 | | *AlgLib.UnequalVarianceTTest |
| 143 | | *AlgLib.OneSampleSignTest |
| 144 | | *AlgLib.WilcoxonSignedRankTest |
| 145 | | *AlgLib.MannWhitneyUtest |
| 146 | | *AlgLib.VarianceChiSquareTest |
| 147 | | *AlgLib.VarianceFTest |
| 148 | | *AlgLib.PearsonCorrelationSignificance |
| 149 | | *AlgLib.SpearmanRankCorrelationSignificance |
| 150 | | *AlgLib.Mann.Whitney.U.test |
| 151 | | ARIMA MLE method Estimate is internally evaluated in local scope to avoid export unexpected symbols |
| 152 | | NameBlok's to call R |
| 153 | | *Rapi: TOL API for R (stdlib/math/R) |
| 154 | | *Rglpk: an interface to the GNU Linear Programing Kit for solving large-scale |
| 155 | | linear programming (LP), mixed integer linear programming (MILP) and |
| 156 | | other related problems (stdlib/math/optim/LP) |
| 157 | | *Rquadprog: Functions to solve Quadratic Programming Problems (stdlib/math/optim/QP) |
| 158 | | *RCoda: output analysis and diagnostics for MonteCarlo Markov Chains (MCMC) |
| 159 | | (stdlib/math/stat/models/bayesian/diagnosis) |
| 160 | | BLR:New strategy simulation that offers a much better efficiency without any loss of |
| 161 | | accuracy or numerical strength (stdlib/math/stat/models/bayesian/linear/_blrc_kernel.tol) |
| | 120 | 1. BSR new blocks: Input and output missing, ARIMA |
| | 121 | 1. Handling with vectorial or time series data in a transparent way |
| | 122 | 1. Non linear filters of output and input |
| | 123 | 1. Handling with normal prior information about parameters of non linear filters. Prior information is given as a log density additive function |
| | 124 | 1. Avoiding to store all nodes to save memory |
| | 125 | 1. Enhanced SQL scripting |
| | 126 | 1. Storing failed factorization matrix |
| | 127 | 1. Allowing fixed sigma in observational nodes |
| | 128 | 1. User can specify the number of simulations to be resumed (do.resume>0) or las simulations to be skiped (do.resume<0) |
| | 129 | 1. Adding SLICE scalar sampler |
| | 130 | 1. New API OneNode to define simple BSR models with just one observational node and optional prior information and order relations |
| | 131 | 1. New non linear filter for probit models |
| | 132 | 1. Ensuring input matrix has changed. Else it will reuse old factorization |
| | 133 | 1. Allowing Inverse-Gamma prior over sigma^2 of BSR segments |
| | 134 | 1. Using faster method LogDens.Almagro.Z_cond_U in draw.ARMS |
| | 135 | 1. New BSR specialization DynHlm for Dynamic Hierarchical Linear Model |
| | 136 | 1. New general API for importing model definition |
| | 137 | 1. New methods BysMcmc::Bsr::Gibbs::Estim and Bsr::Gibbs::Instance(...)::eval |
| | 138 | 1. Adding optional time info assigned to each segment of noise |
| | 139 | 1. Adding and grouping documental information |
| | 140 | 1. New AlgLib Hypothesis testing: |
| | 141 | 1. AlgLib.JarqueBeraTest |
| | 142 | 1. AlgLib.StudentTtest1 |
| | 143 | 1. AlgLib.StudentTtest2 |
| | 144 | 1. AlgLib.UnequalVarianceTTest |
| | 145 | 1. AlgLib.OneSampleSignTest |
| | 146 | 1. AlgLib.WilcoxonSignedRankTest |
| | 147 | 1. AlgLib.MannWhitneyUtest |
| | 148 | 1. AlgLib.VarianceChiSquareTest |
| | 149 | 1. AlgLib.VarianceFTest |
| | 150 | 1. AlgLib.PearsonCorrelationSignificance |
| | 151 | 1. AlgLib.SpearmanRankCorrelationSignificance |
| | 152 | 1. AlgLib.Mann.Whitney.U.test |
| | 153 | stdlib/math/stat/models/bayesian/linear/_blrc_kernel.tol] |
