TF = ismissing (A) returns a logical array, TF, with the same dimensions as A, where true values match the standard missing values in the input data according to their data type.

Standard missing values and their corresponding data types are:

• NaN - for double, single, duration, and calendarDuration arrays.
• NaT - for datetime arrays.
• <missing> - for string arrays.
• <undefined> - for categorical arrays.
• {0x0 char} - for cell arrays of character vectors.

For any data types that do not support missing values, ismissing returns TF = false (size (A)).

Note: the generic ismissing function from the statistics package only operates on core Octave datatypes and it explicitly identifies missing values in double and single arrays, as well as in cell arrays of character vectors. All other data types are handled by the overloaded methods from their respective data class from the datatypes package. Use help class_name.ismissing to find more information about the functional specialization of their respective class implementation.

The optional input indicator can be a scalar or a vector, of the same type as the input data A, specifying alternative missing values in the input data. When specifying indicator values, the standard missing values are ignored, unless explicitly stated in the indicator.

Additional data type matches between indicator and A are:

• double indicators also match single, all integer types, and logical data in A.
• string and char indicators also match categorical data in A.
• char and cellstr indicators also match string data in A.

Note: the generic ismissing function from the statistics package only accepts indicator argument for numeric, logical, and char arrays, as well as for cell arrays of character vectors. For all other core Octave data types, ismissing produces an error. However, indicator is supported for data classes from the datatypes package through their respective class implementation of overloaded methods.

See also: fillmissing, rmmissing, standardizeMissing

Source Code: ismissing