Package index • collinear

Automated Multicollinearity Management

Tools to automatically select sets of variables with a low multicollinearity.

collinear(): Automated multicollinearity management

Variance Inflation Factors

Functions implementing VIF-based methods for multicollinearity filtering.

vif_df(): Variance Inflation Factor

vif_select(): Automated Multicollinearity Filtering with Variance Inflation Factors

Pairwise Correlation

Functions implementing pairwise correlation-based methods for multicollinearity filtering.

cor_clusters(): Hierarchical Clustering from a Pairwise Correlation Matrix

cor_cramer_v(): Bias Corrected Cramer's V

cor_df() cor_numeric_vs_numeric() cor_numeric_vs_categorical() cor_categorical_vs_categorical(): Pairwise Correlation Data Frame

cor_matrix(): Pairwise Correlation Matrix

cor_select(): Automated Multicollinearity Filtering with Pairwise Correlations

Target Encoding

Tools to transform categorical variables into numeric.

target_encoding_lab(): Target Encoding Lab: Transform Categorical Variables to Numeric

target_encoding_mean() target_encoding_rank() target_encoding_loo(): Target Encoding Methods

add_white_noise(): Add White Noise to Encoded Predictor

encoded_predictor_name(): Name of Target-Encoded Predictor

Preference Order

Rank predictors by their association to a response to preserve important ones during multicollinearity filtering.

preference_order(): Quantitative Variable Prioritization for Multicollinearity Filtering

f_auto(): Select Function to Compute Preference Order

f_auto_rules(): Rules to Select Default f Argument to Compute Preference Order

f_functions(): Data Frame of Preference Functions

preference_order_collinear(): Preference Order Argument in collinear()

f_auc_glm_binomial() f_auc_glm_binomial_poly2() f_auc_gam_binomial() f_auc_rpart() f_auc_rf(): Association Between a Binomial Response and a Continuous Predictor

f_r2_pearson() f_r2_spearman() f_r2_glm_gaussian() f_r2_glm_gaussian_poly2() f_r2_gam_gaussian() f_r2_rpart() f_r2_rf(): Association Between a Continuous Response and a Continuous Predictor

f_r2_glm_poisson() f_r2_glm_poisson_poly2() f_r2_gam_poisson(): Association Between a Count Response and a Continuous Predictor

f_v(): Association Between a Categorical Response and a Categorical Predictor

f_v_rf_categorical(): Association Between a Categorical Response and a Categorical or Numeric Predictor

Modelling Tools

Tools to evaluate models, weight cases, and generate model formulas.

case_weights(): Case Weights for Unbalanced Binomial or Categorical Responses

model_formula(): Generate Model Formulas

performance_score_auc(): Area Under the Curve of Binomial Observations vs Probabilistic Model Predictions

performance_score_r2(): Pearson's R-squared of Observations vs Predictions

performance_score_v(): Cramer's V of Observations vs Predictions

Example Data

Real and synthetic datasets used throughout the package examples.

toy: One response and four predictors with varying levels of multicollinearity

vi: Example Data With Different Response and Predictor Types

vi_predictors: All Predictor Names in Example Data Frame vi

vi_predictors_categorical: All Categorical and Factor Predictor Names in Example Data Frame vi

vi_predictors_numeric: All Numeric Predictor Names in Example Data Frame vi

Data Preparation

Internal functions for data preparation and validation.

validate_data_cor(): Validate Data for Correlation Analysis

validate_data_vif(): Validate Data for VIF Analysis

validate_df(): Validate Argument df

validate_encoding_arguments(): Validates Arguments of target_encoding_lab()

validate_predictors(): Validate Argument predictors

validate_preference_order(): Validate Argument preference_order

validate_response(): Validate Argument response

identify_predictors(): Identify Numeric and Categorical Predictors

identify_predictors_categorical(): Identify Valid Categorical Predictors

identify_predictors_numeric(): Identify Valid Numeric Predictors

identify_predictors_type(): Identify Predictor Types

identify_predictors_zero_variance(): Identify Zero and Near-Zero Variance Predictors

identify_response_type(): Identify Response Type

drop_geometry_column(): Removes geometry column in sf data frames