|
|
S+ 8 comprehensive
feature list
S PROGRAMMING LANGUAGE
The award-winning S programming language is at the core of S-PLUS. The
only language created specifically for exploratory data analysis and
statistical modeling, the S programming language allows you to create
statistical applications up to five times faster than with other
languages.
- Object-oriented, interpreted 4GL language
- Interactive exploration and fast prototyping
- Rich data structures: vector, matrix, array, data frame, list
and many more
- User-defined functions, objects, classes, methods and libraries
- Library of over 4000 functions for data manipulation, graphics,
statistical modeling, and integration
- CSAN library of available packages
S-PLUS WORKBENCH DEVELOPMENT ENVIRONMENT
Rapidly create reliable statistical applications with this integrated
development environment for S programmers.
- Based on industry-standard Eclipse framework
- Check-in and check-out files with source code control system
integration
- Intelligent editor for S programs with line numbering, automatic
indentation, and syntax highlighting
- Project, file and task management
- Automatic syntax error detection
- Code outline browser
- Command-line console with history recall
- Object and search path views
- Analytic step-by-step debugger
- Analytic profiling
- Package system for improved porting and deployment
GRAPHICAL USER INTERFACE
A convenient window-based GUI puts common tasks at your fingertips with
easy-to-use menus and dialogs
- File import and export dialogs
- Database import and export dialogs¹
- Dialogs for data preparation, charting and statistical modeling
- Interactive command-line with history recall
- Manage objects with Object Explorer¹
- Script file editor¹
- Multiple data and graphics windows
- Cut-and-paste to Word, PowerPoint and Excel¹
- Integrated Excel spreadsheets¹
- PowerPoint Wizard: quickly create slides from charts¹
- Create custom toolbars, menus and dialogs¹
- On-line help and manuals
- Eclipse based development environment
SCALABLE PIPELINE ARCHITECTURE
Scale statistical applications to gigabytes of data without the need for
additional RAM or 64-bit architectures with this library of data types
and functions for programming with large data sets.
- Data types for out-of-memory vectors, data frames, and time
series
- Use familiar S functions, operators and programming style
- Scalable algorithms for data manipulation, charting and modeling
- High-performance data preparation tools: aggregate, merge, sort,
partition, filter and more
- Data manipulation using built-in SQL processor
- Hexagonal binning plots to explore structure of large data sets
- Scalable model estimation: univariate statistics, linear
regression, analysis of variance, logistic regression, poisson
regression, quasi-likelihood, K‑means clustering, principal
components
- Scalable model scoring for more than 20 model types
GRAPHICAL FUNCTIONS
Explore data and create custom charts with this library of graphical
functions in the S language
- Scatterplots, histograms, pie charts, box plots, bar charts, dot
charts, time series charts, 3-D wireframe charts, image plots and
many more.
- Brush and spin dynamic visualization
- Programmatic control over colors, lines, axes, annotations and
layout
- Unique Trellis™ graphics – create multiple charts conditioned by
levels of one or more variables
- Create interactive, embedded web-based charts with S‑PLUS
Graphlets™
- Element-Specific Graph arguments for plots and command-line
graphics
INTEGRATION
S-PLUS is an open system, designed to integrate with the systems you
already have.
Data and graphics formats
- ASCII: fixed format, comma-separated, and tab-delimited
- Spreadsheets: Excel, Lotus 1-2-3, Quattro Pro
- Application data: SAS 7/8/9, SPSS, Matlab, Minitab, Sigma Plot,
Systat, STATA, Gauss, Epi Info and more
- Database files: Paradox, dBase, Access, FoxPro
- Financial data sources: LIM, Bloomberg, FAME
- Native database clients: SQL Server¹, Oracle, Sybase, IBM DB2
- ODBC interface to compliant databases
- Export graphics as PDF, PostScript, GIF, PNG, JPG, WMF, bitmap,
TIFF and more
APIs and system interfaces
- APIs for C, C++, Java and Fortran
- Language support for pipes, sockets, and files
- DDE, COM and OLE interfaces¹
- XML import and export
- Reporting in XML, PDF, HTML and RTF
STATISTICAL & NUMERICAL TECHNIQUES
S-PLUS is the most comprehensive statistical analysis package available,
and includes all of the following capabilities:
Basic Statistics
- Summary statistics
- Crosstabulations
- Correlation and covariance
- Probabilities, quantiles, densities and random number generation
from many distributions
- Durbin-Watson statistic
Hypothesis Tests and Confidence Intervals
- One-sample and two-sample t-test and Wilcoxon
- Paired t-test
- Correlation: Pearson, Kendall's tau, Spearman's rho
- Goodness-of-Fit: Chi-square, Kolmogorov-Smirnov, Shapiro-Wilk
- Rank tests: Kruskal-Wallis, Friedman
- Proportions: exact Binomial test, Normal approximation
- Contingency tables and tests for independence: Chi-square,
Fisher, Mantel-Haenszel, McNemar
Regression
- Basic linear regression
- Polynomial regression
- Model diagnostics
- Prediction and confidence intervals
- Stepwise selection of models
- Parametric spline models
- Constrained regression
- Logistic regression
- Generalized linear models
Analysis of Variance
- Univariate and multivariate ANOVA
- Flexible specification of variables, covariables, interactions,
nesting, transformations
- Automatic generation of dummy variables
- Choice of contrasts
- Type III sums of squares
- Designed experiments: one-way, two-way, factorial, split-plot,
unbalanced, fractional factorial designs, response surface methods,
robust designs, taguchi methods and more
- Variance component estimation
- Multiple comparisons: Fisher, Tukey, Dunnett, Sidak, Bonferroni,
Scheffé, simulation-based
Nonlinear Regression and Maximum Likelihood
- Nonlinear regression
- Nonlinear maximum likelihood
- Quasi-likelihood
- Constrained nonlinear regression
Nonparametric Regression
- Generalized additive models (GAMs)
- Smoothers: loess, super, kernel, spline
- Projection Pursuit, ACE, and AVAS
Tree Models
- Classification trees
- Regression trees
- Pruning, shrinking, and splitting
- Scoring
Correlated Data Analysis
- Longitudinal data and repeated measures analysis
- Linear (LME), nonlinear (NLME), and generalized mixed effects (GLMM)
models
- Generalized Estimating Equations (GEE)
- Biexponential, first-order compartment, four-parameter logistic
models
- User-defined correlation structures
Resampling
Multivariate Analysis
- Canonical correlation
- Discriminant analysis
- Factor analysis
- Multidimensional scaling
- Principal components
- Biplots
Cluster Analysis
- K-means
- Hierarchical clustering
- Monothetic clustering
- Model-based clustering
- Crisp and fuzzy clustering
- Divisive and agglomerative methods
Quality Control
- Shewhart chart
- Cusum chart
- Charts based on xbar, s, np, p, c, u
Power and Sample Size
- Normal mean
- Binomial proportion
Survival Analysis
- Kaplan-Meier curves
- Cox proportional hazards models with mixed effects
- Left, right, and interval censoring
- Time-dependent covariates and strata
- Multiple event models
- Competing risk models
- Frailty models
- Parametric survival
- Expected survival
- Person years analysis
- Aalen's Additive Regression Model
Time Series Analysis
- Autocovariance, autocorrelation and partial autocorrelation
- Smoothed periodograms
- Box-Jenkins ARIMA models
- Classical and robust AR
- Long-memory models
- Seasonal decompositions
- Fourier transformations
- Classical and robust smoothers and filters
Robust Statistics
- Robust estimation and inferences
- Robust MM regression
- Robust GLM, ANOVA, covariance, principal components, and
discriminant analysis
- Least trimmed squares regression
- Minimum absolute residual regression
- Visually compare robust and traditional methods
Missing Data
- Multiple imputation
- Gaussian, logistic, and conditional Gaussian models
Date, Time, and Calendar Data
- Univariate and multivariate time series
- Aggregation, alignment, merging, and interpolation
- Times and dates from milliseconds to millennia
- Time zones with international daylight savings rules
- Holidays and financial market closures
- Custom time and date formats
- Relative time, time sequence, and event objects
- Powerful time-series charting
Mathematical Computations
- Vector and matrix algebra
- Matrix decompositions
- Systems of linear equations
- Locate roots
- Nonlinear optimization
- Constrained optimization
- Ordinary differential equations
- Numerical integration
ADDITIONAL LIBRARIES
Libraries from Insightful Research and the S‑PLUS user community offer
additional capabilities
- MASS: Modern and Applied Statistics libraries (Venables, Ripley)
included
- Hmisc and Design libraries for biostatistical and epidemiologic
modeling (Harrell) included
- Insightful Research libraries available for download
ADD-ON MODULES
Optional modules add additional capabilities to S+:
- S+FinMetrics: financial econometrics
- S+NuOPT: large-scale constrained optimization
- S+SeqTrial: Clinical trial design and analysis¹
For a quote please
email or call +61 2 92336888 |
RELATED LINKS
|