ismissing

Find missing values

Syntax

TF = ismissing(A)

TF = ismissing(A,indicator)

TF = ismissing(___,'OutputFormat',format)

Description

TF = ismissing(A) returns a logical array that indicates which elements of the input data contain missing values. The size of TF is the same as the size of A.

Missing values are defined according to the data type of A:

• NaN — double, single, duration, and calendarDuration

• NaT — datetime

• <missing> — string

• <undefined> — categorical

• {''} — cell of character vectors

If A is a table, then the data type of each variable defines the missing value for that variable.

For data types with no default definition of a standard missing value, ismissing(A) returns an array or table of logical 0 (false) values the same size as A.

You can use ismissing functionality interactively by adding the Clean Missing Data task to a live script.

TF = ismissing(A,indicator) treats the values in indicator as missing value indicators, ignoring all default indicators listed in the previous syntax. indicator can be a single indicator or multiple indicators. For example, if A is an array of type double, then ismissing(A,[0,-99]) treats 0 and –99 as missing double values instead of NaN.

TF = ismissing(___,'OutputFormat',format) specifies for table or timetable input data the output data type for any of the previous syntaxes.

Examples

NaN Values in Vector

Create a row vector A that contains NaN values, and identify their locations in A.

A = [3 NaN 5 6 7 NaN NaN 9];
TF = ismissing(A)

TF = 1x8 logical array

   0   1   0   0   0   1   1   0

Table with Multiple Data Types

Create a table with variables of different data types.

dblVar = [NaN; 2; 3; 4; 5; 6];
singleVar = single([1; NaN; 3; 4; 5; 6]);
cellstrVar = {'one'; 'two'; ''; 'four'; 'five'; 'six'};
categoryVar = categorical({'red'; 'orange'; 'yellow'; ''; 'blue'; 'indigo'});
dateVar = [datetime(2015,1:4,15) NaT datetime(2015,6,15)]';
stringVar = ["a"; "b"; "c"; "d"; "e"; missing];

A = table(dblVar,singleVar,cellstrVar,categoryVar,dateVar,stringVar)

A=6×6 table
    dblVar    singleVar    cellstrVar    categoryVar      dateVar      stringVar
    ______    _________    __________    ___________    ___________    _________

     NaN           1       {'one'   }    red            15-Jan-2015    "a"      
       2         NaN       {'two'   }    orange         15-Feb-2015    "b"      
       3           3       {0x0 char}    yellow         15-Mar-2015    "c"      
       4           4       {'four'  }    <undefined>    15-Apr-2015    "d"      
       5           5       {'five'  }    blue                   NaT    "e"      
       6           6       {'six'   }    indigo         15-Jun-2015    <missing>

Find the elements with missing values.

ismissing returns an array that has a logical 1 for every corresponding element in A with a missing value. The size of TF is the same as the size of A.

TF = ismissing(A)

TF = 6x6 logical array

   1   0   0   0   0   0
   0   1   0   0   0   0
   0   0   1   0   0   0
   0   0   0   1   0   0
   0   0   0   0   1   0
   0   0   0   0   0   1

Specify Indicators for Missing Values in Table

Create a table where 'NA', '', -99, NaN, and Inf represent missing values. Then, find the elements with missing values.

dblVar = [NaN; 3; Inf; 7; 9];
int8Var = int8([1; 3; 5; 7; -99]);
cellstrVar = {'one'; 'three'; ''; 'NA'; 'nine'};
charVar = ['A'; 'C'; 'E'; ' '; 'I'];

A = table(dblVar,int8Var,cellstrVar,charVar)

A=5×4 table
    dblVar    int8Var    cellstrVar    charVar
    ______    _______    __________    _______

     NaN          1      {'one'   }       A   
       3          3      {'three' }       C   
     Inf          5      {0x0 char}       E   
       7          7      {'NA'    }           
       9        -99      {'nine'  }       I   

Specify the missing value indicators. ismissing returns an array that has a logical 1 for every corresponding element in A with a missing value.

id = {'NA' '' -99 NaN Inf};
TF = ismissing(A,id)

TF = 5x4 logical array

   1   0   0   0
   0   0   0   0
   1   0   1   0
   0   0   1   1
   0   1   0   0

ismissing ignores trailing white space in character arrays. Therefore, because the empty character vector, '', is specified as a missing value indicator, ismissing identifies the empty character vector in A.cellstrVar and also the blank space in A.charVar as missing values.

Input Arguments

A — Input data

vector | matrix | multidimensional array | cell array of character vectors | table | timetable

Input data, specified as a vector, matrix, multidimensional array, cell array of character vectors, table, or timetable.

• If A is a timetable, then ismissing operates on the table data only and ignores NaT and NaN values in the vector of row times.

• If A is a cell array or a table with cell array variables, then ismissing only detects missing elements when the cell array contains character vectors.

indicator — Missing value indicator

scalar | vector | cell array

Missing value indicator, specified as a scalar, vector, or cell array.

• If A is an array, then indicator must be a vector.

• If A is a table or timetable, then indicator can also be a cell array with entries of multiple data types.

The entries of indicator indicate the values that ismissing treats as missing and overrides all default standard missing indicators. If you want to add indicators while maintaining the list of standard indicators, then you must include all default indicators as elements of indicator.

You can use the missing value as an indicator for any missing data represented as NaN, NaT, missing, or <undefined>. If A is a table, then missing is also an indicator for missing cell arrays of character vectors ({''}).

Indicator data types match data types in the entries of A. These are additional data type matches between the indicator and elements of A:

• double indicators match double, single, integer, and logical entries of A.

• string and char indicators, and indicators that are cell arrays of character vectors, match string entries of A.

• string and char indicators match categorical entries of A.

Example: TF = ismissing(A,0) recognizes only 0 as a missing value.

Example: TF = ismissing(A,["Unset" missing]), for categorical array A, recognizes Unset as a missing value in addition to the standard missing value for a categorical value.

Example: TF = ismissing(T,{-99 missing}), for table T, recognizes -99 as a missing value in addition to the standard missing value for the type of each table variable.

format — Output data type

'logical' (default) | 'tabular'

Output data type, specified as one of these values:

• 'logical' — For table or timetable input data, return the output TF as a logical array.

• 'tabular' — For table input data, return the output TF as a table. For timetable input data, return the output TF as a timetable.

For vector, matrix, or multidimensional array input data, OutputFormat is not supported.

Example: ismissing(T,'OutputFormat','tabular')

Tips

• Because integer variables cannot store NaN, use a special integer value (otherwise unused) to indicate missing integer data, such as -99.

• For more information on finding missing strings, see Test for Empty Strings and Missing Values.

• For input data that is a structure array or a cell array of non-character vectors, ismissing returns false. To find missing values in a structure array, apply ismissing to each field in the structure by using the structfun function. To find missing values in a cell array of non-character vectors, apply ismissing to each cell in the cell array by using the cellfun function.

Algorithms

ismissing handles leading and trailing white space differently for indicators that are cell arrays of character vectors, character arrays, or categorical arrays.

• For cell arrays of character vectors, ismissing does not ignore indicator white space. All character vectors must match exactly.

• For character arrays in table variables, ismissing ignores trailing white space in the indicator.

• For categorical arrays, ismissing ignores leading and trailing white space in the indicator.

Alternative Functionality

Live Editor Task

You can use ismissing functionality interactively by adding the Clean Missing Data task to a live script.

Extended Capabilities

Tall Arrays

Calculate with arrays that have more rows than fit in memory.

The ismissing function fully supports tall arrays. For more information, see Tall Arrays.

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

Thread-Based Environment

Run code in the background using MATLAB® backgroundPool or accelerate code with Parallel Computing Toolbox™ ThreadPool.

This function fully supports thread-based environments. For more information, see Run MATLAB Functions in Thread-Based Environment.

Distributed Arrays

Partition large arrays across the combined memory of your cluster using Parallel Computing Toolbox™.

This function fully supports distributed arrays. For more information, see Run MATLAB Functions with Distributed Arrays (Parallel Computing Toolbox).

Version History

Introduced in R2013b

R2022b: Character arrays have no standard missing value

Character arrays have no default definition of a standard missing value. Therefore, ismissing treats blank character array elements (' ') as nonmissing. For example, ismissing(['a'; ' ']) returns logical array [0; 0]. Previously, it returned [0; 1].

To treat blank character array elements as missing, use a missing value indicator to define a blank char as missing. For example, ismissing(['a'; ' '],' ') specifies a blank char as a missing value and returns [0; 1].

R2022a: Return table or timetable containing logical output

For table or timetable input data, return a tabular output TF instead of a logical array by setting the OutputFormat name-value argument to 'tabular'.

R2022a: Accept data types with no standard missing value

The syntax ismissing(A) now returns false (0) when the input data type has no default definition of a standard missing value.

Code that relied on the errors that MATLAB threw for those inputs, such as code within a try/catch block, may no longer catch those errors.