Missing Data in MATLAB

Create and Organize Missing Data

The form that missing values take in MATLAB depends on the data type. For example, numeric data types such asdoubleuseNaN(not a number) to represent missing values.

x = [NaN 1 2 3 4];

You can also use themissingvalue to represent missing numeric data or data of other types, such asdatetime,string, andcategorical. MATLAB automatically converts themissingvalue to the data's native type.

xDouble = [missing 1 2 3 4]

xDouble =1×5NaN 1 2 3 4

xDatetime = [missing datetime(2014,1:4,1)]

xDatetime =1x5 datetimeNaT 01-Jan-2014 01-Feb-2014 01-Mar-2014 01-Apr-2014

xString = [missing"a""b""c""d"]

xString =1x5 string "a" "b" "c" "d"

xCategorical = [missing categorical({'cat1''cat2''cat3''cat4'})]

xCategorical =1x5 categorical<定义> cat1 cat2 cat3 cat4

A data set might contain values that you want to treat as missing data, but are not standard MATLAB missing values in MATLAB such asNaN. You can use thestandardizeMissingfunction to convert those values to the standard missing value for that data type. For example, treat 4 as a missingdoublevalue in addition toNaN.

xStandard = standardizeMissing(xDouble,[4 NaN])

xStandard =1×5NaN 1 2 3 NaN

Suppose you want to keep missing values as part of your data set but segregate them from the rest of the data. Several MATLAB functions enable you to control the placement of missing values before further processing. For example, use the'MissingPlacement'option with thesortfunction to moveNaNs to the end of the data.

xSort = sort(xStandard,'MissingPlacement','last')

xSort =1×51 2 3 NaN NaN

Find, Replace, and Ignore Missing Data

Even if you do not explicitly create missing values in MATLAB, they can appear when importing existing data or computing with the data. If you are not aware of missing values in your data, subsequent computation or analysis can be misleading.

For example, if you unknowingly plot a vector containing aNaNvalue, theNaNdoes not appear because theplotfunction ignores it and plots the remaining points normally.

nanData = [1:9 NaN]; plot(1:10,nanData)

However, if you compute the average of the data, the result isNaN. In this case, it is more helpful to know in advance that the data contains aNaN, and then choose to ignore or remove it before computing the average.

meanData = mean(nanData)

meanData = NaN

One way to findNaNs in data is by using theisnanfunction, which returns a logical array indicating the location of anyNaNvalue.

TF = isnan(nanData)

TF =1x10 logical array0 0 0 0 0 0 0 0 0 1

Similarly, theismissingfunction returns the location of missing values in data for multiple data types.

TFdouble = ismissing(xDouble)

TFdouble =1x5 logical array1 0 0 0 0

TFdatetime = ismissing(xDatetime)

TFdatetime =1x5 logical array1 0 0 0 0

Suppose you are working with a table or timetable made up of variables with multiple data types. You can find all of the missing values with one call toismissing, regardless of their type.

xTable = table(xDouble',xDatetime',xString',xCategorical')

xTable =5×4 tableVar1 Var2 Var3 Var4 ____ ___________ _________ ___________ NaN NaT   1 01-Jan-2014 "a" cat1 2 01-Feb-2014 "b" cat2 3 01-Mar-2014 "c" cat3 4 01-Apr-2014 "d" cat4

TF = ismissing(xTable)

TF =5 x4逻辑阵列1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Missing values can represent unusable data for processing or analysis. Usefillmissingto replace missing values with another value, or usermmissingto remove missing values altogether.

xFill = fillmissing(xStandard,'constant',0)

xFill =1×50 1 2 3 0

xRemove = rmmissing(xStandard)

xRemove =1×31 2 3

Many MATLAB functions enable you to ignore missing values, without having to explicitly locate, fill, or remove them first. For example, if you compute the sum of a vector containingNaNvalues, the result isNaN. However, you can directly ignoreNaNs in the sum by using the'omitnan'option with thesumfunction.

sumNan = sum(xDouble)

sumNan = NaN

sumOmitnan = sum(xDouble,'omitnan')

sumOmitnan = 10