Preprocessing Data

Data cleaning, smoothing, grouping

Data can require preprocessing techniques to ensure accurate, efficient, or meaningful analysis. Data cleaning refers to methods for finding, removing, and replacing bad or missing data. Detecting local extrema and abrupt changes can help to identify significant data trends. Smoothing and detrending are processes for removing noise and polynomial trends from data, while scaling changes the bounds of the data. Grouping and binning methods identify data characteristics by groups.

Live Editor Tasks

Clean Missing Data	Find, fill, or remove missing data in the Live Editor
Clean Outlier Data	Find, fill, or remove outliers in the Live Editor
Find Change Points	Find abrupt changes in data in the Live Editor
Find Local Extrema	Find local maxima and minima in the Live Editor
Smooth Data	Smooth noisy data in the Live Editor
Remove Trends	Remove polynomial trend from data in the Live Editor
Join Tables	Combine two tables using key variables in the Live Editor

Functions

expand all

Missing Data and Outliers

`ismissing`	Find missing values
`rmmissing`	Remove missing entries
`fillmissing`	Fill missing values
`missing`	Create missing values
`standardizeMissing`	Insert standard missing values
`isoutlier`	Find outliers in data
`filloutliers`	Detect and replace outliers in data
`rmoutliers`	Detect and remove outliers in data
`movmad`	Moving median absolute deviation

Detecting Change Points and Local Extrema

`ischange`	Find abrupt changes in data
`islocalmin`	Find local minima
`islocalmax`	Find local maxima

Smoothing and Detrending Data

`smoothdata`	Smooth noisy data
`movmean`	Moving mean
`movmedian`	Moving median
`detrend`	Remove polynomial trend

Normalizing and Scaling Data

`normalize`	Normalize data
`rescale`	Scale range of array elements

Grouping and Binning Data

`discretize`	Group data into bins or categories
`groupcounts`	Number of group elements
`groupfilter`	Filter by group
`groupsummary`	Group summary computations
`grouptransform`	Transform by group
`histcounts`	Histogram bin counts
`histcounts2`	Bivariate histogram bin counts
`findgroups`	Find groups and return group numbers
`splitapply`	Split data into groups and apply function
`rowfun`	Apply function to table or timetable rows
`varfun`	Apply function to table or timetable variables
`accumarray`	Construct array with accumulation