plotEmpiricalCDF

Visualize empirical cumulative distribution function (ecdf) of a variable specified for drift detection

Syntax

plotEmpiricalCDF(DDiagnostics)

plotEmpiricalCDF(DDiagnostics,Variable=variable)

plotEmpiricalCDF(ax,___)

St = plotEmpiricalCDF(___)

Description

plotEmpiricalCDF(DDiagnostics) plots the ecdf values of the baseline and target data for the continuous variable with the lowest p-value. If there are no continuous variables in the data, then plotEmpiricalCDF does not generate a plot and returns a warning.

If you set the value of EstimatePValues to false in the call to detectdrift, then plotEmpiricalCDF displays NaN for the p-value and the drift status.

plotEmpiricalCDF(DDiagnostics,Variable=variable) plots the ecdf for the variable specified by variable.

plotEmpiricalCDF(ax,___) plots into axes ax instead of gca.

St = plotEmpiricalCDF(___) plots the ecdf and returns an array of Stair objects in St. Use this to inspect and adjust the properties of the ecdf. To learn more about the properties of the Stair object, see Stair Properties.

Examples

Plot ECDF for Variable with Lowest p-Value

Generate baseline and target data with three variables, where the distribution parameters of the second and third variables change for target data.

rng('default') % For reproducibility
baseline = [normrnd(0,1,100,1),wblrnd(1.1,1,100,1),betarnd(1,2,100,1)];
target = [normrnd(0,1,100,1),wblrnd(1.2,2,100,1),betarnd(1.7,2.8,100,1)];

Perform permutation testing for all variables to check for any drift between the baseline and target data.

DDiagnostics = detectdrift(baseline,target)

DDiagnostics = 
  DriftDiagnostics

              VariableNames: ["x1"    "x2"    "x3"]
       CategoricalVariables: []
                DriftStatus: ["Stable"    "Drift"    "Warning"]
                    PValues: [0.3850 0.0050 0.0910]
        ConfidenceIntervals: [2x3 double]
    MultipleTestDriftStatus: "Drift"
             DriftThreshold: 0.0500
           WarningThreshold: 0.1000


  Properties, Methods

Plot the ecdf for the variable with the lowest p-value.

plotEmpiricalCDF(DDiagnostics)

plotEmpiricalCDF by default plots the ecdf of the baseline and target data for the variable with the lowest p-value, which, in this case, is variable x2. You can see the difference between the two empirical cumulative distribution functions. It also displays the p-value and the drift status for variable x2.

Plot ECDF for Specified Variable

Generate baseline and target data with three variables, where the distribution parameters of the second and third variables change for target data.

rng('default') % For reproducibility
baseline = [normrnd(0,1,100,1),wblrnd(1.1,1,100,1),betarnd(1,2,100,1)];
target = [normrnd(0,1,100,1),wblrnd(1.2,2,100,1),betarnd(1.7,2.8,100,1)];

Perform permutation testing for all variables to check for any drift between the baseline and target data.

DDiagnostics = detectdrift(baseline,target)

DDiagnostics = 
  DriftDiagnostics

              VariableNames: ["x1"    "x2"    "x3"]
       CategoricalVariables: []
                DriftStatus: ["Stable"    "Drift"    "Warning"]
                    PValues: [0.3850 0.0050 0.0910]
        ConfidenceIntervals: [2x3 double]
    MultipleTestDriftStatus: "Drift"
             DriftThreshold: 0.0500
           WarningThreshold: 0.1000


  Properties, Methods

Plot the ecdf for the third variable.

plotEmpiricalCDF(DDiagnostics,Variable="x3")

plotEmpiricalCDF plots the ecdf for baseline and target data. It also displays the estimated p-value and the drift status for the same variable.

Plot ECDF for Variables in Tiled Layout

Load the sample data.

load humanactivity

For details on the data set, enter Description at the command line. Assign the first 250 observations for columns 10 to 15 as baseline data and next 250 as target data.

baseline = feat(1:250,10:15);
target = feat(251:500,10:15);

Test for drift on all variables.

DDiagnostics = detectdrift(baseline,target)

DDiagnostics = 
  DriftDiagnostics

              VariableNames: ["x1"    "x2"    "x3"    "x4"    "x5"    "x6"]
       CategoricalVariables: []
                DriftStatus: ["Drift"    "Stable"    "Stable"    ...    ]
                    PValues: [1.0000e-03 0.5080 0.2370 1.0000e-03 0.5370 ... ]
        ConfidenceIntervals: [2x6 double]
    MultipleTestDriftStatus: "Drift"
             DriftThreshold: 0.0500
           WarningThreshold: 0.1000


  Properties, Methods

Drift statuses for variables x4 and x6 are drift and warning, respectively. Plot the empirical cumulative distribution function values for x4 and x6 in a tiled layout.

tiledlayout(1,2);
ax1 = nexttile;
plotEmpiricalCDF(DDiagnostics,ax1,Variable="x4")
ax2= nexttile;
plotEmpiricalCDF(DDiagnostics,ax2,Variable="x6")

There is a greater difference between the ecdf of baseline and target data for variable x4, for which detectdrift detected the shift.

Input Arguments

DDiagnostics — Diagnostics of the permutation testing for drift detection

DriftDiagnostics object

Diagnostics of the permutation testing for drift detection, specified as a DriftDiagnostics object returned by detectdrift.

variable — Variable for which to visualize ecdf

string | character vector | integer index

Variable for which to visualize the ecdf, specified as a string, a character vector, or an integer index.

Example: Variable="x3"

Example: Variable=3

Data Types: single | double | char | string

ax — Axes to plot into

Axes object | UIAxes object

Axes for plotEmpiricalCDF to plot into, specified as an Axes or UIAxes object. If you do not specify ax, then plotEmpiricalCDF creates the plot using the current axes. For more information on creating an axes object, see axes and uiaxes.

Version History

Introduced in R2022a