multivariate
JMulTi 4.1
JMulTi is a Time Series Analysis with Java more>> JMulTi is a Time Series Analysis with Java
JMulTi was originally created as a tool for certain econometric procedures in time series analysis that are especially difficult to use and that are not available in other packages, like Impulse Response Analysis with bootstrapped confidence intervals for
Now many other features have been integrated as well to make it possible to convey a comprehensive analysis. Limitations of this software can be overcome by exporting datasets or computation results and use them with other programs. For an overview of the underlying software concept, see the JStatCom page.
Main features:
- various tools for creating, transforming, editing time series
- Unit Root tests: ADF, HEGY (quarterly, monthly), Schmidt-Phillips, KPSS, Unit Root test with structural break
- Cointegration tests: Johansen Cointegration test with response surfaces, Saikkonen & L?tkepohl test
- kernel density estimation
- spectral density plots
- crossplots
- autocorrelation analysis
- VAR modelling (with arbitrary deterministic/exogenous variables)
- subset model estimation
- output in matrix form
- automatic model selection (various strategies based on information criteria)
- residual analysis with tests for nonnormality, autocorrelation, ARCH, spectrum, kernel density, autocorrelation plots, crosscorrelation
- GARCH analysis for residuals
- Impulse Responses with bootstrapped confidence intervals also for accumulated responses, orthogonal and forecast error versions
- Forecast Error Variance Decomposition
- forecasting, also levels from 1st differences, asymptotic confidence intervals for levels
- causality tests
- stability analysis: bootstrapped Chow tests, recursive parameters, recursive residuals, CUSUM test
- SVAR modelling: AB model, Blanchard-Qua Model with bootstrapped standard errors
- SVAR Forecast Error Variance Decomposition
- SVAR Impulse Responses with bootstrapped confidence intervals
- VECM modelling (with arbitrary deterministic/exogenous variables)
- restrictions on cointegration space, Wald test for beta restrictions
- Johansen, Two Stage, S2S estimation procedures
- EC term can be fully or partly predetermined
- subset model estimation
- output in matrix form
- automatic model selection (various strategies based on information criteria)
- residual analysis with tests for nonnormality, autocorrelation, ARCH, spectrum, kernel density, autocorrelation plots, crosscorrelation
- Impulse Responses with bootstrapped confidence intervals also for accumulated responses, orthogonal and forecast error versions
- Forecast Error Variance Decomposition
- forecasting, also levels from 1st differences, asymptotic confidence intervals for levels
- causality tests
- stability analysis: bootstrapped Chow tests, recursive parameters, recursive eigenvalues
- SVEC modelling with bootstrapped standard errors
- SVEC Forecast Error Variance Decomposition
- SVEC Impulse Responses with bootstrapped confidence intervals
- univariate ARCH, GARCH, T-GARCH estimation with different error distributions
- residual analysis for ARCH residuals with robustified test for no remaining ARCH (S. Lundbergh, T. Teraesvirta), plotting of variance process, kernel density for residuals
- multivariate GARCH(1,1) estimation, residual analysis, plotting of variance process together with univariate estimates, kernel density for residuals
- lag selection for univariate models based on linear and nonlinear selection criteria
- nonlinear estimation with configurable 3D plots
- residual analysis for estimation residuals
- model selection for volatility process
- estimation of volatility process
- residual analysis for volatility estimation residuals
Antaeus 2.3.57
Antaeus is a powerful and flexible system for finding patterns in multivariate data. more>>
Antaeus 2.3.57 is a powerful and flexible system for finding patterns in multivariate data. It is intended to be used by a single investigator working from individual data tables.
Major Features:
- It provides tools for looking at your data when you don't yet know what you're looking for, led by your intuition and knowledge about the emergent phenomena represented.
- Antaeus does not use and requires no knowledge of statistics. However, it is based on the functional use of fundamental mathematical principles. The basic plot type used everywhere in
- Antaeus is the scatter plot, supplemented by its high-density counterpart, the sunflower plot. These plots are supported by quantile plots and histograms, which are used to see the data of a single variable. Scatter plots are used to look for the relationship between two variables, but as the number of variables increases, more and more scatter plots become necessary to visualize all the possible interrelationships.
- Antaeus is built on a framework to which increasingly specialized data visualization tools will be attached. For instance, our current 2.3 release builds upon our previous 2.2 release with the addition systems for adding new measures which are functions of existing measures, adding new dimensions from the partitioning of scatter plots, the scatter plot array, the scatter plot track, single pixel data points (for very large data sets), Z-space view for creating subsets from slices or regions, sampling options, and the ability to stop the drawing of any plot (for large data sets). As more tools and subsystems are attached to the framework, so will the depth with which Antaeus can be used to look for patterns.
- All the plots generated by Antaeus are very highly finished and do not require any user interaction to achieve this level of quality.
- Users can size the plots by dragging the edges of their windows and they have extensive control over the palettes used to color them, but all their construction details are handled proactively.
- Antaeus is also a platform from which plots may be published. Any plot may be saved to file or clipboard as a Windows Enhanced Metafile (EMF), which can be embedded in reports, papers, and presentations created in Microsoft Office products such as Word, PowerPoint and Publisher. Support for EMF is also increasing among non-Microsoft publication suites. These sophisticated programs are empowered by the use of plots they cannot possibly construct themselves.
Enhancements:
- Fix
-
- Brushes were not being drawn in cubes that contain no dimensions.
- This oversight has been corrected.
- Change: Extra error handling has been added to help us track down the lock-out problem noted under build 2.3.56.
Requirements: Windows Vista and XP
Janina Ratkowska Group - A system for viewing and gaining insight into multivariate data through the use of high-quality, interrelated plots. Antaeus. Antaeus 2.2 BuildLicense:Freeware
Use of subsets as brush in multivariate plots. Magnification of measures. Reversal of measure scale
JSpline+ 1.26
JSpline+ is designed to support numerically intensive calculations in Java wi... more>> JSpline+ is designed to support numerically intensive calculations in Java with matrices and splines.Contains classes for univariate and multivariate spline approximation on scattered meshes, as well as core matrix and linear system solution classes and data access<<less
AMRandom 4.1
Generate Random Numbers to fit Probability Distribution - Full Delphi Source more>>
The following random number generators are included in AMRandom:
- Normal (Gaussian)
- Gamma
- Chi-squared
- Exponential
- Weibull
- Beta
- t
- Multivariate Normal
- Generalized inverse Gaussian
- Binomial (2 different ones)
- Negative Binomial
- von Mises
- Cauchy
Includes full Delphi Source and Demo.
System requirements:
- Borland Delphi 5 or better
) - Gamma - Chi-squared - Exponential - Weibull - Beta - t - Multivariate Normal - Generalized inverse Gaussian - Binomial (2 different ones) - Negative Binomial - von Mises - Cauchy Includes fullOrbital 1.1.8
Orbital is a Java class library providing object-oriented representations and algorithms for logic, mathematics, and AI more>>
Generally speaking, the conceptual idea behind the Orbital library is to provide extensional services and components for Java, which surround the heart of many scientific applications.
Hence the name Orbital library. In order to satisfy the requirements of high reusability, the design of this foundation class library favors flexibility, conceptual simplicity and generalisation. So many sophisticated problems can be solved easily with its adaptable components.
Especially useful tools are functor composition tools, including their integration into numerical and symbolic mathematics, as well as general functional evaluation schemes, and several algorithmic templates including search and planning. There are implementations of different logics and automated theorem proving systems, and computer algebra system routines.
Main features:
- The automatic theorem prover jImp and its reusable components for set of support and ordered resolution with clause indexing, subsumption, tautology elimination as well as Davis-Putnam-Loveland inference.
- The computer algebra system (CAS) components jAbr with Groebner-bases, gcd, symbolic differentiation, and data representation of vectors, matrices, multivariate polynomials, real numbers, rationals, integers, complex numbers, quotients, etc.
- Several solution algorithms for search and planning problems.
- Machine learning algorithms, including evolutionary algorithms.
Classifion 1.5
Classifion is software for classification of substances using their mass-spectra more>> Classifion is chemometrics software for classification of substances using their mass-spectra. The used discriminant analysis is based on Principal Component Analysis with Mahalanobis Distance (PCA-MD) - part of multivariate analysis. Classifion has been tested using mass-spectra, but there is no limitation to be used with any other kind of characteristic spectra (e.g. VIS, IR, NMR).<<less
FreeFore 6
FreeFore does as it sounds, it is a free software version of Autobox 6.0 where you can test it out on 720 classic textbook time series. more>>
FreeFore 6 does as it sounds, it is a free software version of Autobox 6.0 where you can test it out on 720 classic textbook time series. You can also use your data, but you are restricted to getting one period out forecast. You can see the model and the value. Autobox was the best performing automated software in the 2008 Neural Network Forecasting Competition!
Instead of using a pick best approach Autobox automatically builds the model (univariate and multivariate) all while identifying and incorporating any interventions. It selects the best lead/lag structures for each input series and corrects for omitted variables (holidays or price changes that have affected the historical data, but that the system has no knowledge of) by identifying pulses, seasonal pulses, level shifts and local time trends, and then adding the needed structure through surrogate variables.
Enhancements :
- Handles daily data specific issues like day of the week, week of the year, and day of the month effects.
It builds Box-Jenkins models for univariate and multivariate data and produces a forecast. You can use your data and or data supplied with the product with graphs and spreadsheet style outputBioGoggles 1.0.0
Multivariate statistics and visualization of gene and protein expression data more>> BioGoggles provides easy analysis and visualization of your multivariate data. Automatic calculation of p-values and ROC areas for each variable. Scatter, Box-Whisker and ROC plots. Principal Component Analysis and 3D plot<<less
ForeStock 1.1
Collection of advanced forecasting algorithms for Equis Metastock(R) more>> Ever growing collection of advanced forecasting algorithms. Currently implemented:
ARIMA with automatic best parameter set search and smart seasonal trend detection.
Stepwise multivariate regression with the selection of the best regression subset over the whole bar history.
Pattern profitability expert picks best profitable patterns by comparing current market profile with historical profiles.
Finite State Markov Automation provides Markov process emulation of market states with dynamically adjusted transitional probabilities
Proven backhistory. Model is recalculated on each historical time step using only those data which were available prior to actual historical trade date, so you see performance exactly same as you would observe on real trades.
Full intraday capable solutions available on the synchronous with datastream basis.
Real-time speed in model training. Models have accuracy comparable to most advanced neural network solvers while retaining the training speed allowing recalculate on the fly.
Ready to use professional style Expert Advisors, Indicators and System Testers
Open scripts for for writing your own trading strategies using direct algorithmic calls.
Full automated installer integrated with Equis Metastock versions 7, 8 and 9. Just plug-and-play.
Free 2 week trial - you just start earn money and pay for the product out of your profits.
And much, much more new features and algorithms to come.<<less
- Page: 1 of 1
- 1
