nanmean

(Not recommended) Mean, ignoring NaN values

nanmean is not recommended. Use the MATLAB^® function mean instead. With the mean function, you can specify whether to include or omit NaN values for the calculation. For more information, see Compatibility Considerations.

Syntax

y = nanmean(X)

y = nanmean(X,'all')

y = nanmean(X,dim)

y = nanmean(X,vecdim)

Description

y = nanmean(X) returns the mean of the elements of X, computed after removing all NaN values.

• If X is a vector, then nanmean(X) is the mean of all the non-NaN elements of X.

• If X is a matrix, then nanmean(X) is a row vector of column means, computed after removing NaN values.

• If X is a multidimensional array, then nanmean operates along the first nonsingleton dimension of X. The size of this dimension becomes 1 while the sizes of all other dimensions remain the same. nanmean removes all NaN values.

For information on how nanmean treats arrays of all NaN values, see Tips.

y = nanmean(X,'all') returns the mean of all elements of X, computed after removing NaN values.

y = nanmean(X,dim) returns the mean along the operating dimension dim of X, computed after removing NaN values.

y = nanmean(X,vecdim) returns the mean over the dimensions specified in the vector vecdim. The function computes the means after removing NaN values. For example, if X is a matrix, then nanmean(X,[1 2]) is the mean of all non-NaN elements of X because every element of a matrix is contained in the array slice defined by dimensions 1 and 2.

Examples

Mean of Matrix Data

Find the column means for matrix data with missing values.

X = magic(3);
X([1 6:9]) = NaN

X = 3×3

   NaN     1   NaN
     3     5   NaN
     4   NaN   NaN

y = nanmean(X)

y = 1×3

    3.5000    3.0000       NaN

Mean of All Values

Find the mean of all the values in an array, ignoring missing values.

Create a 2-by-5-by-3 array X with some missing values.

X = reshape(1:30,[2 5 3]);
X([10:12 25]) = NaN

X = 
X(:,:,1) =

     1     3     5     7     9
     2     4     6     8   NaN


X(:,:,2) =

   NaN    13    15    17    19
   NaN    14    16    18    20


X(:,:,3) =

    21    23   NaN    27    29
    22    24    26    28    30

Find the mean of the elements of X.

y = nanmean(X,'all')

y = 15.6538

Mean Along Scalar Dimension

Find the row means for matrix data with missing values by specifying to compute the means along the second dimension.

X = magic(3);
X([1 6:9]) = NaN

X = 3×3

   NaN     1   NaN
     3     5   NaN
     4   NaN   NaN

y = nanmean(X,2)

y = 3×1

     1
     4
     4

Mean Along Vector Dimension

Find the mean of a multidimensional array over multiple dimensions.

Create a 2-by-5-by-3 array X with some missing values.

X = reshape(1:30,[2 5 3]);
X([10:12 25]) = NaN

X = 
X(:,:,1) =

     1     3     5     7     9
     2     4     6     8   NaN


X(:,:,2) =

   NaN    13    15    17    19
   NaN    14    16    18    20


X(:,:,3) =

    21    23   NaN    27    29
    22    24    26    28    30

Find the mean of each page of X by specifying dimensions 1 and 2 as the operating dimensions.

ypage = nanmean(X,[1 2])

ypage = 
ypage(:,:,1) =

     5


ypage(:,:,2) =

   16.5000


ypage(:,:,3) =

   25.5556

For example, ypage(1,1,1) is the mean of the non-NaN elements in X(:,:,1).

Find the mean of the elements in each X(i,:,:) slice by specifying dimensions 2 and 3 as the operating dimensions.

yrow = nanmean(X,[2 3])

yrow = 2×1

   14.5385
   16.7692

For example, yrow(2) is the mean of the non-NaN elements in X(2,:,:).

Input Arguments

X — Input data

scalar | vector | matrix | multidimensional array

Input data, specified as a scalar, vector, matrix, or multidimensional array.

If X is an empty array, then nanmean(X) is NaN. For more details, see Tips.

Data Types: single | double

dim — Dimension to operate along

positive integer scalar

Dimension to operate along, specified as a positive integer scalar. If you do not specify a value, then the default value is the first array dimension whose size does not equal 1.

dim indicates the dimension whose length reduces to 1. size(y,dim) is 1 while the sizes of all other dimensions remain the same.

Consider a two-dimensional array X:

• If dim is equal to 1, then nanmean(X,1) returns a row vector containing the mean for each column.

• If dim is equal to 2, then nanmean(X,2) returns a column vector containing the mean for each row.

If dim is greater than ndims(X) or if size(X,dim) is 1, then nanmean returns X.

Data Types: single | double

vecdim — Vector of dimensions

positive integer vector

Vector of dimensions, specified as a positive integer vector. Each element of vecdim represents a dimension of the input array X. The output y has length 1 in the specified operating dimensions. The other dimension lengths are the same for X and y.

For example, if X is a 2-by-3-by-3 array, then nanmean(X,[1 2]) returns a 1-by-1-by-3 array. Each element of the output is the mean of the elements on the corresponding page of X.

Data Types: single | double

Output Arguments

y — Mean values

scalar | vector | matrix | multidimensional array

Mean values, returned as a scalar, vector, matrix, or multidimensional array.

Tips

• When nanmean computes the mean of an array of all NaN values, the array is empty once the NaN values are removed and, therefore, the sum of the remaining elements is 0. Because the mean calculation involves division by 0, the mean value is NaN. The output NaN is not a mean of NaN values.

Compatibility Considerations

nanmean is not recommended

Not recommended starting in R2020b

nanmean is not recommended. Use the MATLAB function mean instead. There are no plans to remove nanmean.

To update your code, change instances of the function name nanmean to mean. Then specify the 'omitnan' option for the nanflag input argument.

The mean function has these advantages over the nanmean function:

• mean offers more extended capabilities for supporting tall arrays, GPU arrays, distribution arrays, C/C++ code generation, and GPU code generation.

• mean returns an output value with a specified data type.

Extended Capabilities

C/C++ Code Generation

Generate C and C++ code using MATLAB® Coder™.

Usage notes and limitations:

• The 'all' and vecdim input arguments are not supported.

• The dim input argument must be a compile-time constant.

• If you do not specify the dim input argument, the working (or operating) dimension can be different in the generated code. As a result, run-time errors can occur. For more details, see Automatic dimension restriction (MATLAB Coder).

For more information on code generation, see Introduction to Code Generation and General Code Generation Workflow.

GPU Arrays

Accelerate code by running on a graphics processing unit (GPU) using Parallel Computing Toolbox™.

Usage notes and limitations:

• The 'all' and vecdim input arguments are not supported.

For more information, see Run MATLAB Functions on a GPU (Parallel Computing Toolbox).