Statistics and Machine Learning Toolbox

R2026b
Analyze and model data using statistics and machine learning

Statistics and Machine Learning Toolbox™ provides functions and apps for statistical analysis and machine learning in MATLAB®.

For statistical analysis, you can use descriptive statistics to explore data, fit probability distributions, test hypotheses, perform analysis of variance (ANOVA), plan experiments, validate measurement systems, and monitor processes. You can use functions for programmatic analysis and interactive apps for guided workflows such as distribution fitting, design of experiments (DOE), and Gage R&R.

For machine learning, you can train regression, classification, clustering, and other models interactively or programmatically. The toolbox supports feature engineering, dimensionality reduction, model interpretation, Simulink® integration, and C/C++ code generation for deployment. For an interactive experience, you can use the Classification Learner or Regression Learner apps to explore data, select features, choose validation schemes, tune hyperparameters, and evaluate multiple algorithms side-by-side.

Get Started

Learn the basics of Statistics and Machine Learning Toolbox

Descriptive Statistics and Visualization

Data import and export, descriptive statistics, visualization

Probability Distributions and Hypothesis Tests

Data frequency models, random sample generation, parameter estimation, and hypothesis testing

Industrial Statistics

Design of experiments (DOE), survival and reliability analysis, statistical process control

ANOVA

Analysis of variance and covariance, multivariate ANOVA, repeated measures ANOVA

Regression

Linear, generalized linear, nonlinear, and nonparametric techniques for supervised learning

Classification

Supervised and semi-supervised learning algorithms for binary and multiclass problems

Cluster Analysis and Anomaly Detection

Unsupervised learning techniques to find natural groupings, patterns, and anomalies in data

Dimensionality Reduction and Feature Extraction

PCA, factor analysis, feature selection, feature extraction, and more

Machine Learning Pipelines

Execute and deploy end-to-end machine learning workflows as pipelines

Simulink and Code Generation

Simulate machine learning models using Simulink and generate C/C++ code

Figure depicting the results of training a Gaussian process regression model for predicting battery state of charge. This figure comes from the example, "Predict Battery State of Charge Using Machine Learning".

Applications

Apply statistics and machine learning methods to industry-specific workflows