Features

S Programming Language

The award-winning S programming language is at the core of Spotfire S+. 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+ 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-PLUSGraphlets™
• Element-Specific Graph arguments for plots and command-line graphics

Integration

S-PLUSis 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-PLUSis 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

• Bootstrap
• Jackknife

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-PLUSuser 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

Optional modules add additional capabilities to S-PLUS:

• S+FinMetrics: financial econometrics
• S+NuOPT: large-scale constrained optimization
• S+SeqTrial: Clinical trial design and analysis¹

Supported Platforms

• Windows 2000,Windows XP, Windows Vista
• Sun Solaris (SPARC)
• Red Hat and SUSE Linux (Intel)

• 

¹ Windows Only

Feature List

Statistics

• Statistical Summaries and Tests
• Extreme Value Theory
• Copula Modeling and Estimation

Time Series Tools

• Complete suite of Date and Calendar Time Series Objects
• Aggregation and Disaggregation
• Missing Value Interpolation
• Technical Indicators
• Intra-day Moving Average
• Factor ARCH models, Engle et al
• Matrix/matrix models

Econometric Estimation

• Generalized Method of Moments
• Efficient Method of Moments
• Linear and nonlinear SUR
• Vector Autoregressive Models (VARs)
• Bayesian VARs
• Vector Error Correction Models

Complex Dynamic Models

• Long memory models
• State space models
• Nonlinear regime switching models

Strategies

• Rolling Estimation and Backtesting
• Multifactor Models
• Fixed Income Analysis

• 

Feature List

• Complete set of powerful and efficient solvers included
• Very fast and robust optimization
• Optimize problems with thousands of variables and constraints
• Complete flexibility in specifying objective function and constraints
• Specify problems in a natural way using SIMPLE, a class of objects in the S language
• Automated symbolic differentiation of objective function
• Complete control over solver tolerance and options

Optimization Problems Solved

• Linear Programming
• Mixed-Integer Linear Programming
• General Convex Programming
• Convex Quadratic Programming
• General Non-Linear Programming
• Mixed Integer Quadratic Programming

Optimization Methods

• Primal-dual interior point method with higher-order correction for linear programming
• Simplex method for linear programming
• Primal-dual interior point method based on line search for general convex Programming models including convex quadratic programming models
• Primal-dual interior point method based on trust region for general non-linear programming models
• Primal-dual interior point method based on quasi-Newton method for general non-linear programming models
• Active set method for convex quadratic programming models and mixed integer quadratic programming models

SIMPLE: A set of classes for defining optimization problems

• "Set" class: element, union, intersection, difference, direct product
• Defining objective functions: expressions, variables, integer variables, parameters.
• Defining contraints: relational, equality, inequality, inclusion in set, conditional expressions.

• 

Feature List

Improved Study Design

• Reduce costs and bring drugs to market sooner with sequential studies
• Interactive interface makes it easy to explore tradeoffs between designs

Comprehensive Design and Evaluation Tools

• Sequential designs in unified family of Kittelson and Emerson, including all commonly used group sequential designs
• Family of designs based on error spending function of Lan and DeMets
• Comprehensive evaluation tools, including power, conditional power, sample size distribution, inference at the boundaries, and Bayesian analyses

Completely Extensible

• Powerful S language for extending functionality to fit your needs; from small to large projects, from simple to complex analyses

Flexible Monitoring Methods

• Implementation of stopping rules based on error spending functions or constrained boundaries

Easily Analyze and Interpret Your Results

• Exact p-values
• Exact confidence intervals
• Bias adjusted estimates of treatment effect

Easy to Learn

• State-of-the art GUI (Graphical User Interface)
• Documentation designed for the clinical trialist
• Fully integrated with S-PLUSdata analysis software

Advanced Visualization Tools

• Comprehensive set of specialized plots for designing studies, including power curves, ASN (average sample size) curves, and stopping probabilities
• Trellis graphics for powerful and effective comparison of designs

Powerful Validation Techniques

• Results validated against standard designs in the group sequential literature

• 

In This Section

• Business Intelligence
• Quality Analysis & Six Sigma
• Risk Analysis & Simulation
• Statistical Modeling
  • TIBCO Spotfire S+
  • TIBCO Spotfire S+ Modules
  • TIBCO Spotfire Miner
  • TIBCO Spotfire S+ Server
  • TIBCO Spotfire Clinical Graphics
  • Brochures
    • What's New in TIBCO Spotfire S+ 8.1
    • TIBCO Spotfire S+ 8.1
    • TIBCO Spotfire S+ 8.1 Feature List
    • TIBCO Spotfire Miner
    • TIBCO Spotfire Miner Feature List
    • Analytics in Action
    • TIBCO Spotfire Clincal Graphics
    • FinMetrics
    • S+ NuOpt
    • S+ SeqTria
  • White Papers
    • a
    • b
    • c
• Optimization
• Econometrics
• Flowcharting
• Product Training
• Data Analysis Services
• Product Resources

• Contact for proposal
• Road map for new customer