Data Transforms and Smoothing
Directly integrated into the Wolfram Language's uniform architecture for handling lists of data is an array of highly optimized algorithms for transforming and smoothing datasets that can routinely involve millions of elements.
Rescale ▪ Clip ▪ Normalize ▪ Standardize ▪ Accumulate ▪ Differences
Threshold — adaptively determine a suitable threshold
MovingAverage — find the simple moving average with any span
ExponentialMovingAverage — find the exponential moving average with damping
MovingMedian — find the moving median with any span
MovingMap — map a function over a moving window with any span
ArrayFilter — map a function over a moving window in an array of any depth
Interpolation — find an interpolation of any order in any number of dimensions
Fit — linear least-squares fit
FindFit — find a constrained nonlinear fit to data
ListConvolve, ListCorrelate — convolve or correlate data with any kernel
ListDeconvolve — restore convolved data
CellularAutomaton — apply a cellular automaton rule in any number of dimensions
Fourier, InverseFourier — discrete Fourier transform and inverse
Filters »
GaussianFilter ▪ LaplacianFilter ▪ WienerFilter ▪ MedianFilter ▪ ...
Wavelet Analysis »
DiscreteWaveletTransform ▪ WaveletThreshold ▪ ...
Outliers & Missing Data
DeleteAnomalies — learn from data to delete anomalous elements
SynthesizeMissingValues — fill in missing values by imputing from existing data
Peak Analysis
FindPeaks — find the positions of peaks in data
EstimatedBackground — estimate a smooth background in data