Descriptive Statistics

Range, central tendency, standard deviation, variance, correlation

Descriptive statistics quantitatively describe features of a sample of data, such as the basic mean or standard deviation. Cumulative methods report a statistic as you move through the elements of an array. Moving methods report a statistic within a local window of array elements, then move to the next window.

Functions

expand all

Basic Statistics

`min`	Minimum elements of array
`mink`	Find `k` smallest elements of array
`max`	Maximum elements of array
`maxk`	Find `k` largest elements of array
`bounds`	Minimum and maximum values of an array
`mean`	Average or mean value of array
`median`	Median value of array
`mode`	Most frequent values in array
`std`	Standard deviation
`var`	Variance
`rms`	Root mean square value

Percentiles and Quantiles

`prctile`	Percentiles of data set
`quantile`	Quantiles of data set
`iqr`	Interquartile range of data set

Forecasting Metrics

`rmse`	Root mean squared error between arrays (Since R2022b)
`mape`	Mean absolute percentage error between arrays (Since R2022b)

Covariance and Correlation

`cov`	Covariance
`corrcoef`	Correlation coefficients
`xcov`	Cross-covariance
`xcorr`	Cross-correlation

Cumulative Statistics

`cummax`	Cumulative maximum
`cummin`	Cumulative minimum

Moving Statistics

`movmin`	Moving minimum
`movmax`	Moving maximum
`movmean`	Moving mean
`movmedian`	Moving median
`movmad`	Moving median absolute deviation
`movstd`	Moving standard deviation
`movvar`	Moving variance
`movsum`	Moving sum
`movprod`	Moving product

Data Summary

summary Data summary

Probability Distributions

kde Kernel density estimate for univariate data (Since R2023b)

Topics

Computing with Descriptive Statistics
Analyze data with basic statistics.
Linear Correlation
Covariance and correlation coefficients help to describe the linear relationship between variables.
Linear Regression
Least squares fitting is a common type of linear regression that is useful for modeling relationships within data.
Interactive Fitting
The Basic Fitting UI is an interactive data modeling tool.
Programmatic Fitting
There are many functions in MATLAB^® that are useful for data fitting.

Featured Examples

Identify and Visualize Correlated Variables

Calculate, filter, rank, and visualize correlations between variables.

Open Live Script