Predictive Modeling Absolute Mean Difference Cutoff for Continuous Predictors Absolute Proportion Difference Cutoff for Class Predictors Add reflected y variable to binary model Algorithm Apply adaptive weights Asymmetric Loss Evaluation Proportion Asymmetric Loss Fitting Proportion Automated Model Type Average Pool Size Binary Dependent Variable Event Value Block Partition Variable Boosted Fraction of Training Observations in a Single Tree Boosted Maximum Number of Iterations Boosted or Forest Random Number Seed Boosted Shrinkage Factor Censor Variable Centroid or Distance Summarization Method Change output folder in Settings moved to right panel Class Covariates Color Variable Color Variables Combined Output Data Set Name Compute Individual Model Fits Continuous Covariates Continuous Variables Correlation Radius for Clustering Criterion for Stopping Model Selection Current Review Custom Costs Custom Prior Probabilities CV Partitioning Method Data Columns to be Transposed Data Step Statements Dependent Class Variable Dependent Variable Dependent Variables Distance Metric for Analysis Distance Metric Distribution Early Stopping Threshold Elastic Net L2 Penalty Experimental Design Data Set Experimental Design SAS Data Set Feature Selection Criterion Filter for BLUP Coefficients Filter for PLS Coefficients Filter to Include Observations Filter to Include Predictor Class Variables Filter to Include Predictor Continuous Variables Fixed Test Data Set Fixed Test Set Folder of CVMC Results 1 Folder of CVMC Results 2 Folder of CVMC Results 3 Folder of CVMC Results 4 Folder of CVMC Results 5 Folder of Predictive Modeling Settings Folder of Settings Files Folder of Test Data Sets Folder of Test Sets For Nominal Dependent Variables, compute distances to: Forest Max Number of Variables to Compute Predictor Importance Forest Max Number of Variables to Consider for Splitting a Node Forest Maximum Number of Trees Forest: Output filtered predictors list Generate HTML Output Group Variable Grouping Variable Heterogeneous Variance Components Hold-Out Method Hold-Out Size, Specify as: ID Variable Impute zeros for missing continuous predictor values Impute zeros for missing values Include 3D plots Inner Loop Algorithm Input SAS Data Set Input Tall Data Set Input Wide Data Set Iteration End Number Iteration Start Number JMP Script Output File Name K for K-Fold CV K for K-Fold or 1/K Hold-Out Kernel Function Kernel Function for Computing Posterior Probabilities L for Leave-L-Out L1 Regularization Parameter L2 Regularization Parameter Label Variable Link Function List every model fit List SAS Output for all model fits List-Style Specification of Continuous Variables List-Style Specification of Data Columns to be Transposed List-Style Specification of Lock-In Class Predictor Variables List-Style Specification of Lock-In Continuous Predictor Variables List-Style Specification of Predictor Class Variables List-Style Specification of Predictor Continuous Variables List-Style Specification of Variables to Be Standardized Lock-In Class Predictor Variables Lock-In Continuous Predictor Variables -log10(p-Value) Cutoff log10 Regularization Parameter Max Number of Categories Allowed in a Predictor Max Number of Effects in the Model Max Number of Variables to Consider for Splitting a Node Maximum Depth of Tree Maximum Number of Buckley-James Iterations Maximum Number of Filtered Predictors Maximum Number of Grid Nodes to Use Maximum Number of K-Means Clusters / Predictors Maximum Number of Predictors to Use Maximum Number of Steps Maximum Number of Trees Maximum Number of Variables to Select with Model Averaging Maximum Number of Variables to Select with Pooling Maximum Order of Interactions Maximum Size of Training Set Maximum Time for Area under Survival Curves Method Metric Minimum Number of Observations Required for a Branch Minimum Number of Observations Required for a Categorical Value Minimum Size of Training Set Minimum Time for Area under Survival Curves Mode Model Selection Method Multiple Testing Method Nominalize Continuous Dependent Variables Number of BLUPs to Use in Prediction Number of Generations Number of Grid Points for Each Learning Curve Number of model averaging samples Number of Nearest Neighbors Number of PLS Components Number of Predictor Variables to Select Number of Predictors Included in Model Number of Principal Components Number of Random Hold-Out Iterations Number of Random Iterations Number of Rounds of Selection Number of Rows in Input Data to Use in Test Run Number of Subsets Containing a Particular Variable Numerical Parameter for Advanced Standardization Methods Optimize the continuous predictor set Outer Loop Test Set Output Data Set Output Data Set Name Output Data Set Prefix Output Experimental Design Data Set Output Folder Output Tall Data Set Output Wide Data Set P for P-Percent-Hold-Out Perform Buckley-James Estimation Perform Cross Validation Model Comparison Perform Learning Curve Model Comparison Perform model averaging Perform poolwise selection Perform recoding with respect to: Plot weights for predictor variables Predictor Class Variables Predictor Continuous Variables Prefix for Tall Column Names Prefix for Wide Column Names Primary Input SAS Data Set Primary Output Data Set Name Prior Probabilities / Prevalences PROC DISCRIM Options PROC DISTANCE Options PROC GENESELECT Statement Options PROC GLIMMIX Additional Statements PROC GLIMMIX Class Variables PROC GLIMMIX Fixed Effects PROC GLIMMIX MODEL Statement Options PROC GLIMMIX Statement Options PROC GLMSELECT Modeling Options PROC GLMSELECT Statement Options PROC LIFEREG Modeling Options PROC LIFEREG Statement Options PROC LOGISTIC MODELING Options PROC LOGISTIC Response Options PROC LOGISTIC Statement Options PROC MIXED Additional Statements PROC MIXED Class Variables PROC MIXED Fixed Effects PROC MIXED MODEL Statement Options PROC MIXED Statement Options PROC PHREG Modeling Options PROC PHREG Statement Options PROC PLS Statement Options PROC QUANTSELECT Modeling Options PROC QUANTSELECT Statement Options Quantile Level Quantile Level for Quantile Regression Random Number Seed Random Number Seed for Forest Reference Time for Comparing Survival Curves Reference Times for Comparing Survival Curves Regard missing values as valid for prediction Root Mean Square Error Convergence Tolerance Rule Mix Maximum Number of Initial Rules Rule Mix Maximum Number of Secondary Rules Secondary Input SAS Data Set Secondary Output Data Set Name Separate Bar Charts for Each Model Separate Bar Charts for Each Test Set Separate Charts for Each Model Separate Charts for Each Test Set Server Output Directory Settings Files Settings for Which to Construct Learning Curves Settings to Cross Validate Settings to Use for Test Data Set Evaluation Settings to Use for Test Set Evaluation Significance Level for Adding Variables Significance Level for Keeping Variables Significance Level for Retaining Variables Similarity Measure SL for Adding Variables SL for Keeping Variables SL for Retaining Variables Standardization Method for Predictor Continuous Variables Standardize Predictors Row-Wise Statistical Testing Method for Continuous Predictors Study Test Data Set Test Data Sets Test Sets Time to Event Variable Transformation for Predictor Continuous Variables Type of Dependent Variable Usage of K-Means Clusters Use Forest to create interaction indicators Use Forest to filter predictors Use Grid Computing Use K-Means clustering to reduce predictors Use Leave-One-Out error rate as the Fitness function Use life regression to filter predictors Use PROC HPLOGISTIC Use simple Cox Proportional Hazards function to filter predictors Use statistical testing to filter predictors Validation Data Set Value of the Censor Variable that Indicates Censoring Values of the Censor Variable that Indicate Censoring Variable Selection Method Variables Defining Tall Column Names Variables Defining Wide Column Names Variables to Be Standardized Weight for Censored Observations Weight Variable Weighting Function Where Clause for Subsetting Input Data in Test Run