Function reference • tidyfit

Basic functionality

The functions regress() and classify() provide basic modeling wrappers for tidy data.

regress: Linear regression on tidy data

classify: Classification on tidy data

Generic methods

The ‘tidyfit.models’ frame produced by ‘regress()’ or ‘classify()’ can be used to obtain coefficients or predictions.

coef(<tidyfit.models>): Extract coefficients from a tidyfit.models frame

predict(<tidyfit.models>): Predict using a tidyfit.models frame

fitted(<tidyfit.models>): Obtain fitted values from models in a tidyfit.models frame

residuals(<tidyfit.models>): Obtain residuals from models in a tidyfit.models frame

Modeling engine

The model wrapper m() standardizes a large number of linear regression and classification algorithms.

m(): Generic model wrapper for tidyfit

Implemented algorithms

.model(<adalasso>): Adaptive Lasso regression or classification for tidyfit

.model(<bayes>): Bayesian generalized linear regression for tidyfit

.model(<blasso>): Bayesian Lasso regression for tidyfit

.model(<bma>): Bayesian model averaging for tidyfit

.model(<boost>): Gradient boosting regression for tidyfit

.model(<bridge>): Bayesian ridge regression for tidyfit

.model(<chisq>): Pearson's Chi-squared test for tidyfit

.model(<cor>): Pearson's correlation for tidyfit

.model(<enet>): ElasticNet regression or classification for tidyfit

.model(<gets>): General-to-specific regression for tidyfit

.model(<glm>): Generalized linear regression for tidyfit

.model(<glmm>): Generalized linear mixed-effects model for tidyfit

.model(<hfr>): Hierarchical feature regression for tidyfit

.model(<lasso>): Lasso regression and classification for tidyfit

.model(<lm>): Linear regression for tidyfit

.model(<mrmr>): Minimum redundancy, maximum relevance feature selection for tidyfit

.model(<mslm>): Markov-Switching Regression for tidyfit

.model(<pcr>): Principal Components Regression for tidyfit

.model(<plsr>): Partial Least Squares Regression for tidyfit

.model(<quantile>): Quantile regression for tidyfit

.model(<relief>): ReliefF and RReliefF feature selection algorithm for tidyfit

.model(<rf>): Random Forest regression or classification for tidyfit

.model(<ridge>): Ridge regression and classification for tidyfit

.model(<robust>): Robust regression for tidyfit

.model(<subset>): Best subset regression and classification for tidyfit

.model(<svm>): Support vector regression or classification for tidyfit

.model(<tvp>): Bayesian Time-Varying Regression for tidyfit

Data

tidyfit comes with a built-in financial factor returns dataset.

Factor_Industry_Returns: Industry-Factor Returns Data Set