NMath Stats Changelog

------------------------------------------------------------------------------
 Version 3.4.0 
------------------------------------------------------------------------------
- Simplified the NMath Stats installer. Updated license keys will be issued
  at upgrade time. A license file must now be deployed with your NMath Stats
  applications.
- Upgraded to Intel MKL 10.3 Update 6 with resulting performance increases.
- Added assembly NMathStatsChartMicrosoft.dll containing class NMathStatsChart,
  which provides static methods for plotting NMath Stats types using the
  Microsoft Chart Controls for .NET. (For more information, see whitepaper
  "NMath Stats Visualization Using the Microsoft Chart Controls.")
- Modified class TDistribution to accept fractional degrees of freedom.
- Added class TwoSampleUnpairedUnequalTTest. Unlike TwoSampleUnpairedTTest,
  a pooled estimate of the variance is not used.


------------------------------------------------------------------------------
 Version 3.3.0 
------------------------------------------------------------------------------
- Upgraded to Intel MKL 10.3 Update 2 with resulting performance increases.
- Added class PearsonsChiSquareTest for performing a chi-square hypothesis test.


------------------------------------------------------------------------------
 Version 3.2.0 
------------------------------------------------------------------------------
- Upgraded to Intel MKL 10.2 Update 5 with resulting performance increases.
- Added classes InputVariableCorrelator and ReducedVarianceInputCorrelator
  to induce a desired rank correlation among a set of random input variables. 
- Added class KruskalWallisTest for performing a Kruskal-Wallis rank sum test. 
- Added class JohnsonDistribution to encapsulate the Johnson system of
  distributions.
- Added classes GoodnessOfFit and GoodnessOfFitParameter for testing the
  goodness of fit of least squares model-fitting classes, such as
  LinearRegression, PolynomialLeastSquares, and OneVariableFunctionFitter.
- Added class BoxCox for computing the Box-Cox power tranformations.
- Added StatsFunction.Median() for float types.
- Fixed issue where StatsFunctions.SumOfSquares() could occasionally return a
  negative value.
- Improved support for missing values in DataFrame.Load().


------------------------------------------------------------------------------
 Version 3.1.0 
------------------------------------------------------------------------------
- Upgraded to Intel MKL 10.2 Update 1.
- Added class KMeansClustering for k-means clustering. The k-means clustering
  method assigns data points into k groups such that the sum of squares from
  points to the computed cluster centers is minimized. 
- Added class MovingWindowFilter for 1D data filtering. Static class methods
  are provided for generating coefficients to implement a moving average
  filter and a Savitzky-Golay smoothing filter. 
- Added method Cronbach() to class StatsFunctions for calculating the
  standardized Cronbach's alpha test for reliability. 
- Added class NMFNonsmoothUpdate, an NMF update algorithm that minimizes a
  cost functional designed to explicitly represent sparseness, in the form
  of non-smoothness.


------------------------------------------------------------------------------
 Version 3.0.0 
------------------------------------------------------------------------------
- Upgraded to Intel MKL 10.1 Update 2.
- Added class NMFClustering for performing data clustering using iterative
  nonnegative matrix factorization (NMF). 
- Added class NMFConsensusMatrix for computing a consensus matrix from among
  multiple NMF clusterings. 
- Added classes ConnectivityMatrix and OrderedConnectivityMatrix for
  representing connectivity matrices.
- Changed abbreviation for nonnegative matrix factorization from NNMF to NMF in
  all class and method names.   
- Added single precision support for principle component analysis. Renamed class
  PrincipleComponentAnalysis to DoublePCA, and added FloatPCA.  
- Fixed bug in LinearRegression where default IRegressionCalculation was not
  thread-safe.
- Prevented duplicate column name in DataFrame. Duplicate names are now
  automatically suffixed with "_1", "_2",...,"_n".
- Added new overloads for StatsFunction.CovarianceMatrix() for FloatMatrix and
  StatsFunction.Covariance() for FloatVector.
- Added custom visualizer for DataFrame.
- Organized classes into solution folders inside Visual Studio.
- Added VS 2008 solutions for all examples and source editions.
- Added Add(System.Object) method to all types which implement IEnumerable.