Statistics and Machine Learning Toolbox™ contains a variety of functions that work with tall arrays. Tall arrays provide a convenient way to work with data that does not fit in memory, that is, the sample size can be arbitrarily large. To create a tall array, first create a datastore that references the data, and then use the tall function to convert the datastore into a tall array. For more information about tall arrays in MATLAB^®, see Tall Arrays (MATLAB). For a list of supported statistics functions, see Tall Array Support, Usage Notes, and Limitations.

If you have Parallel Computing Toolbox™, then the use of parallel computing can speed up certain statistical computations with tall arrays. To use parallel computing with tall arrays, see Extend Tall Arrays with Other Products (MATLAB).