Data Analysis for Social Sciences in GAUSS

Main Applications of GAUSS in Social Sciences

GAUSS software provides a complete set of tools for social science analytics. Whether you're just getting started with data collection or finalizing results, GAUSS has the data analytics tools you need.

Whatever your area of research, GAUSS supports all your data analysis needs, large or small.

FieldExample applications
Political Science
  • Event count models
  • Modeling voting behavior
  • Applied policy evaluation
  • Game theory modeling
Healthcare economics
  • Panel count models
  • Healthcare expenditure modeling
  • Truncated and censored discrete choice models
  • Average treatment effects
  • Meta-analysis
  • Bootstrap resampling and repeated measure designs
  • Goodness-of-fit testing for psychometrics
  • Between-groups hypothesis testing
Transportation Studies
  • Commuting choice models
  • Activity based models
  • Demand management and forecasting
  • Mode choice models
  • Free flow speed estimation

Regression Models, Time Series Models, and Other Main Functions of GAUSS for Social Scientists

GAUSS covers a comprehensive set of data analysis tools from data organization and management to advanced panel data techniques.

Data cleaning, processing, and management

General Regression Analysis

Pre-built GAUSS functions can be used to efficiently and intuitively implement fundamental regression models including:

Time Series Analysis

With GAUSS time series analysis is made easy and efficient whether you're just getting started or developing new cutting edge methods. GAUSS time series capabilities include:

Discrete Choice Analysis

GAUSS provides a full suite of tools for analyzing qualitative choice models found across many social sciences. GAUSS's discrete choice tools cover everything from binary and multinomial models to logistic regression.

  • Multinomial logit models
  • Logistic regression modeling
    • L2 and L1 regularized classifiers
    • L2 and L1-loss linear support vector machines (SVM)
  • Model selection and assessment tools
    • Full model and restricted model log-likelihoods
    • Chi-square statistics
    • Agresti’s G-squared statistic
    • McFadden’s pseudo-R-squared statistic
    • Madella’s pseudo-R-squared statistic
    • Akaike information criterion (AIC)
    • Bayesian information criterion (BIC)
    • Likelihood ratio statistics and accompanying probability values
    • Cragg and Uhler’s normed likelihood ratios
    • Count and adjusted count R-squared

Panel Data Analysis

  • Data aggregation and within-group statistics
  • Panel data unit root tests
    • Breitung and Das panel unit root test
    • Im, Pesaran, and Shin (IPS) panel unit root test
    • Levin-Lin-Chu (LLC) panel unit root test
    • Pesaran unit root test in the presence of cross-section dependence
    • Modified CADF and CIPS panel unit root tests
    • Bai and Ng PANIC panel unit root test
    • Harris and Tzavalis panel unit root test
    • Hadri panel data unit root tests
    • Panel unit root tests with structural breaks
    • Im, Lee, & Tieslau panel LM unit root test with level shifts
    • Lee and Tieslau panel LM unit root test with level and trend shifts
    • Nazlioglu & Karul panel stationarity test with gradual structural shifts
  • One-way individual effects
    • One-way fixed effects
    • One-way random effects
    • Least squares pooled ols
    • Least squares dummy variable (LSDV)
  • Cross dependence tests
    • Pesaran test for cross-dependence
    • Friedman test for cross-dependence
    • Frees test for cross-dependence
  • Causality tests
    • Granger causality
    • Toda & Yamamoto causality test
    • Single Fourier-frequency Granger causality test
    • Single Fourier-frequency Toda & Yamamoto causality test
    • Cumulative Fourier-frequency Granger causality test
    • Cumulate Fourier-frequency Today & Yamamoto test
    • Fischer testing for Granger causality in heterogeneous mixed panels
    • Zhnc and Zn test statistics for Granger non-causality in heterogeneous panels
    • Panel SUR Wald statistics
  • Model diagnostics and assessment tests
    • Hausman test for specification
    • Lagrange multiplier test for error components model

GAUSS Applications Designed for Social Sciences


Time series MT

Includes comprehensive tools for time series data analysis including
  • MLE and state-space estimation
  • Unit root and cointegration testing
  • Model diagnostics and forecasting
  • Nonlinear time series models

Linear regression MT

Provides procedures for estimating single equations or systems of equations including:
  • Two-stage least squares.
  • Three-stage least squares.
  • Seemingly unrelated regression.

Discrete Choice

Provides an adaptable, efficient, and user-friendly environment for linear data classification including
  • Binary and count models.
  • Multinomial logit models.
  • Logistic regression.
  • Tools for model selection and assessment.

Bayesian Estimation Tools

Provides a suite of pre-built tools for Bayesian estimation and analysis including:
  • Data generation.
  • Markov-chain Monte Carlo estimation (MCMC).
  • Full post-estimation graphing and reporting.
  • Maximum likelihood estimation (MLE) initialization.

Maximum Likelihood MT

Provides a suite of flexible, efficient and trusted tools for the solution of the maximum likelihood problem with bounds on the parameters. Includes:
  • Variety of descent and line search algorithms.
  • Analytical and numerical derivatives.
  • Dynamic algorithm switching.
  • Multiple tools for statistical inference.

Constrained Maximum Likelihood MT

Provides a suite of flexible, efficient and trusted tools for the solution of the maximum likelihood problem with general constraints on the parameters. Features include:
  • Linear and nonlinear equality and inequality constraints.
  • Trust region method.
  • A variety of descent and line search algorithms.
  • Analytical and numerical derivatives.
  • Dynamic algorithm switching.
  • Multiple methods for statistical inference.

Optimization MT

Optimization MT provides tools for efficient optimization including:
  • Select descent algorithms.
  • Step-length methods.
  • Algorithm switching.

Constrained Optimization MT

Solves the nonlinear programming problem, subject to general constraints on the parameters. Includes:
  • Linear or nonlinear constraints.
  • Equality or inequality constraints.
  • Uses sequential quadratic programming method in combination with several descent methods.
  • Trust region method.

Descriptive Statistics MT

Provides basic statistics for the variables in GAUSS datasets. These statistics describe and test univariate and multivariate features of the data and provide information for further analysis.

Algorithmic Derivatives

Provides tools for computing algorithmic derivatives.
  • Works independently of other applications.
  • Can be used with any application that needs derivatives.
  • The use of algorithmic derivatives can improve accuracy and convergence speeds.

Industries that use GAUSS Data Analysis Tools

Social scientists across a wide range of industries use GAUSS. GAUSS is found in

  • Universities
  • Government agencies
  • Non-governmental organizations
  • Nonprofit research organizations
  • Corporations

Whether analyzing election results, modeling transportation mode choices, or constructing event count models, GAUSS offers the tools you need to succeed.

Icons of some organizations where GAUSS is used.

Benefits of GAUSS for Social Scientists

GAUSS provides a fast and flexible environment for data analysis. Whether you are performing ordinary least squares regressions or developing cutting-edge algorithms, GAUSS provides tangible advantages including:

  • Over 1000 pre-built statistical functions.
  • Light-weight and efficient analytics engine designed to make the most of your hardware and provide optimized computation speed.
  • Intuitive matrix-based programming language for transparent and easy to understand programming.
  • Fully interactive environment for speeding up your workflow from exploring data to analyzing results.
  • Comprehensive documentation and examples.
  • Comprehensive data support including CSV, Excel HDF5, SAS, Stata, text delimited files.
  • Relational database support including MySQL, PostgreSQL, SQLite, Microsoft SQL Server, Oracle, IBM DB2, HBase, Hive and MongoDB.

Compatibility of GAUSS with Other Software

GAUSS is built to seamlessly integrate into any analytics environment:

  • GAUSS is fully compatible with SAS, STATA, HDF5, CSV, and Excel datasets.
  • Efficiently connect powerful analytics to any internal or customer-facing data source, application, or interface with the GAUSS Engine.
  • Full technical support for assistance when migrating from and integrating with other software platforms.

Have a Specific Question?

Get a real answer from a real person

Need Support?

Get help from our friendly experts.