• Documentation
    • About ​ValidMind
    • Get Started
    • Guides
    • Support
    • Releases

    • Python Library
    • ValidMind Library

    • ValidMind Academy
    • Training Courses
  • Log In
    • Public Internet
    • ValidMind Platform · US1
    • ValidMind Platform · CA1

    • Private Link
    • Virtual Private ValidMind (VPV)

    • Which login should I use?
  1. tests
  2. model_validation
  3. RegressionFeatureSignificance
  • ValidMind Library Python API

  • Python API
  • 2.8.25
  • init
  • init_dataset
  • init_model
  • init_r_model
  • get_test_suite
  • log_metric
  • preview_template
  • print_env
  • reload
  • run_documentation_tests
  • run_test_suite
  • tags
  • tasks
  • test
  • log_text
  • experimental_agent
  • RawData
    • RawData
    • inspect
    • serialize

  • Submodules
  • __version__
  • datasets
    • classification
      • customer_churn
      • taiwan_credit
    • credit_risk
      • lending_club
      • lending_club_bias
    • nlp
      • cnn_dailymail
      • twitter_covid_19
    • regression
      • fred
      • lending_club
  • errors
  • test_suites
    • classifier
    • cluster
    • embeddings
    • llm
    • nlp
    • parameters_optimization
    • regression
    • statsmodels_timeseries
    • summarization
    • tabular_datasets
    • text_data
    • time_series
  • tests
    • data_validation
      • ACFandPACFPlot
      • ADF
      • AutoAR
      • AutoMA
      • AutoStationarity
      • BivariateScatterPlots
      • BoxPierce
      • ChiSquaredFeaturesTable
      • ClassImbalance
      • CommonWords
      • DatasetDescription
      • DatasetSplit
      • DescriptiveStatistics
      • DickeyFullerGLS
      • Duplicates
      • EngleGrangerCoint
      • FeatureTargetCorrelationPlot
      • Hashtags
      • HighCardinality
      • HighPearsonCorrelation
      • IQROutliersBarPlot
      • IQROutliersTable
      • IsolationForestOutliers
      • JarqueBera
      • KPSS
      • LJungBox
      • LaggedCorrelationHeatmap
      • LanguageDetection
      • Mentions
      • MissingValues
      • MissingValuesBarPlot
      • MutualInformation
      • PearsonCorrelationMatrix
      • PhillipsPerronArch
      • PolarityAndSubjectivity
      • ProtectedClassesCombination
      • ProtectedClassesDescription
      • ProtectedClassesDisparity
      • ProtectedClassesThresholdOptimizer
      • Punctuations
      • RollingStatsPlot
      • RunsTest
      • ScatterPlot
      • ScoreBandDefaultRates
      • SeasonalDecompose
      • Sentiment
      • ShapiroWilk
      • Skewness
      • SpreadPlot
      • StopWords
      • TabularCategoricalBarPlots
      • TabularDateTimeHistograms
      • TabularDescriptionTables
      • TabularNumericalHistograms
      • TargetRateBarPlots
      • TextDescription
      • TimeSeriesDescription
      • TimeSeriesDescriptiveStatistics
      • TimeSeriesFrequency
      • TimeSeriesHistogram
      • TimeSeriesLinePlot
      • TimeSeriesMissingValues
      • TimeSeriesOutliers
      • TooManyZeroValues
      • Toxicity
      • UniqueRows
      • WOEBinPlots
      • WOEBinTable
      • ZivotAndrewsArch
      • nlp
    • model_validation
      • AdjustedMutualInformation
      • AdjustedRandIndex
      • AutoARIMA
      • BertScore
      • BleuScore
      • CalibrationCurve
      • ClassifierPerformance
      • ClassifierThresholdOptimization
      • ClusterCosineSimilarity
      • ClusterPerformanceMetrics
      • ClusterSizeDistribution
      • CompletenessScore
      • ConfusionMatrix
      • ContextualRecall
      • CumulativePredictionProbabilities
      • DurbinWatsonTest
      • FeatureImportance
      • FeaturesAUC
      • FowlkesMallowsScore
      • GINITable
      • HomogeneityScore
      • HyperParametersTuning
      • KMeansClustersOptimization
      • KolmogorovSmirnov
      • Lilliefors
      • MeteorScore
      • MinimumAccuracy
      • MinimumF1Score
      • MinimumROCAUCScore
      • ModelMetadata
      • ModelParameters
      • ModelPredictionResiduals
      • ModelsPerformanceComparison
      • OverfitDiagnosis
      • PermutationFeatureImportance
      • PopulationStabilityIndex
      • PrecisionRecallCurve
      • PredictionProbabilitiesHistogram
      • ROCCurve
      • RegardScore
      • RegressionCoeffs
      • RegressionErrors
      • RegressionErrorsComparison
      • RegressionFeatureSignificance
      • RegressionModelForecastPlot
      • RegressionModelForecastPlotLevels
      • RegressionModelSensitivityPlot
      • RegressionModelSummary
      • RegressionPerformance
      • RegressionPermutationFeatureImportance
      • RegressionR2Square
      • RegressionR2SquareComparison
      • RegressionResidualsPlot
      • RobustnessDiagnosis
      • RougeScore
      • SHAPGlobalImportance
      • ScoreProbabilityAlignment
      • ScorecardHistogram
      • SilhouettePlot
      • TimeSeriesPredictionWithCI
      • TimeSeriesPredictionsPlot
      • TimeSeriesR2SquareBySegments
      • TokenDisparity
      • ToxicityScore
      • TrainingTestDegradation
      • VMeasure
      • WeakspotsDiagnosis
      • sklearn
      • statsmodels
      • statsutils
    • prompt_validation
      • Bias
      • Clarity
      • Conciseness
      • Delimitation
      • NegativeInstruction
      • Robustness
      • Specificity
      • ai_powered_test
  • unit_metrics
  • vm_models

On this page

  • RegressionFeatureSignificance
    • Purpose
    • Test Mechanism
    • Signs of High Risk
    • Strengths
    • Limitations
  • Edit this page
  • Report an issue
  1. tests
  2. model_validation
  3. RegressionFeatureSignificance

validmind.RegressionFeatureSignificance

RegressionFeatureSignificance

@tags('statistical_test', 'model_interpretation', 'visualization', 'feature_importance')

@tasks('regression')

defRegressionFeatureSignificance(model:validmind.vm_models.VMModel,fontsize:int=10,p_threshold:float=0.05):

Assesses and visualizes the statistical significance of features in a regression model.

Purpose

The Regression Feature Significance metric assesses the significance of each feature in a given set of regression model. It creates a visualization displaying p-values for every feature of the model, assisting model developers in understanding which features are most influential in their model.

Test Mechanism

The test mechanism involves extracting the model's coefficients and p-values for each feature, and then plotting these values. The x-axis on the plot contains the p-values while the y-axis denotes the coefficients of each feature. A vertical red line is drawn at the threshold for p-value significance, which is 0.05 by default. Any features with p-values to the left of this line are considered statistically significant at the chosen level.

Signs of High Risk

  • Any feature with a high p-value (greater than the threshold) is considered a potential high risk, as it suggests the feature is not statistically significant and may not be reliably contributing to the model's predictions.
  • A high number of such features may indicate problems with the model validation, variable selection, and overall reliability of the model predictions.

Strengths

  • Helps identify the features that significantly contribute to a model's prediction, providing insights into the feature importance.
  • Provides tangible, easy-to-understand visualizations to interpret the feature significance.

Limitations

  • This metric assumes model features are independent, which may not always be the case. Multicollinearity (high correlation amongst predictors) can cause high variance and unreliable statistical tests of significance.
  • The p-value strategy for feature selection doesn't take into account the magnitude of the effect, focusing solely on whether the feature is likely non-zero.
  • This test is specific to regression models and wouldn't be suitable for other types of ML models.
  • P-value thresholds are somewhat arbitrary and do not always indicate practical significance, only statistical significance.
RegressionErrorsComparison
RegressionModelForecastPlot
  • ValidMind Logo
    ©
    Copyright 2025 ValidMind Inc.
    All Rights Reserved.
    Cookie preferences
    Legal
  • Get started
    • Model development
    • Model validation
    • Setup & admin
  • Guides
    • Access
    • Configuration
    • Model inventory
    • Model documentation
    • Model validation
    • Model workflows
    • Reporting
    • Monitoring
    • Attestation
  • Library
    • For developers
    • For validators
    • Code samples
    • API Reference
  • Training
    • Learning paths
    • Courses
    • Videos
  • Support
    • Troubleshooting
    • FAQ
    • Get help
  • Community
    • Slack
    • GitHub
    • Blog
  • Edit this page
  • Report an issue