Statistics and Machine Learning Toolbox™ provides two additional data types. Work with ordered and
unordered discrete, nonnumeric data using the nominal
and
ordinal
data types. Store multiple variables,
including those with different data types, into a single object using the
dataset
array data type. However, these data types
are unique to Statistics and Machine Learning Toolbox. For greater cross-product compatibility, use the categorical
or table
data types,
respectively, available in MATLAB®. For more information see Create Categorical Arrays,
Create and Work with Tables, or watch Tables and Categorical Arrays.
dataset | (Not Recommended) Arrays for statistical data |
Nominal and ordinal arrays store data that have a finite set of discrete levels, which might or might not have a natural order.
Advantages of Using Nominal and Ordinal Arrays
Easily manipulate category levels, carry out statistical analysis, and reduce memory requirements.
Grouping variables are utility variables used to group or categorize observations.
Dummy variables let you adapt categorical data for use in classification and regression analysis.
Other MATLAB Functions Supporting Nominal and Ordinal Arrays
Learn about MATLAB functions that support nominal and ordinal arrays.
Create Nominal and Ordinal Arrays
Create nominal and ordinal arrays using nominal
and
ordinal
, respectively.
Categorize numeric data into a categorical ordinal array using
ordinal
.
Change the labels for category levels in nominal or ordinal arrays using
setlabels
.
Add and drop levels from a nominal or ordinal array.
Merge categories in a nominal or ordinal array using
mergelevels
.
Reorder the category levels in nominal or ordinal arrays using
reorderlevels
.
Determine sorting order for ordinal arrays.
Plot data grouped by the levels of a categorical variable.
Summary Statistics Grouped by Category
Compute summary statistics grouped by levels of a categorical variable.
Test Differences Between Category Means
Test for significant differences between category (group) means using a t-test, two-way ANOVA (analysis of variance), and ANOCOVA (analysis of covariance) analysis.
Index and Search Using Nominal and Ordinal Arrays
Index and search data by its category, or group.
Linear Regression with Categorical Covariates
Perform a regression with categorical covariates using categorical arrays and
fitlm
.
Dataset arrays store data with heterogeneous types.
Create a Dataset Array from Workspace Variables
Create a dataset array from a numeric array or heterogeneous variables existing in the MATLAB workspace.
Create a Dataset Array from a File
Create a dataset array from the contents of a tab-delimited or a comma-separated text, or an Excel file.
Add and delete observations in a dataset array.
Add and delete variables in a dataset array.
Access Data in Dataset Array Variables
Work with dataset array variables and their data.
Select Subsets of Observations
Select an observation or subset of observations from a dataset array.
Sort Observations in Dataset Arrays
Sort observations (rows) in a dataset array using the command line.
Merge dataset arrays using join
.
Stack or Unstack Dataset Arrays
Reformat dataset arrays using stack
and unstack
.
Find, clean, and delete observations with missing data in a dataset array.
Calculations on Dataset Arrays
Perform calculations on dataset arrays, including averaging and summarizing with a grouping variable.
Export a dataset array from the MATLAB workspace to a text or spreadsheet file.
Dataset Arrays in the Variables Editor
The MATLAB Variables editor provides a convenient interface for viewing, modifying, and plotting dataset arrays.
Index and Search Dataset Arrays
Learn the many ways to index into dataset arrays.
Regression Using Dataset Arrays
This example shows how to perform linear and stepwise regression analyses using dataset arrays.