Example dataset | Creating a gbm distribution object | Setting training parameters | Additional variable and fitting parameters | Cross Validation | Parallelisation | Putting it all together | Default behaviour | Identifying the optimal iteration | Fitting additional trees | Predictions on other data | Relative influence of predictors | Plotting the marginal effects of selected variables | Additional useful functions

Distributions with model specific parameters | Cox proportional hazards model | Pairwise distribution | Quantile | TDist | Tweedie

Distributions with model specific parameters | Cox proportional hazards model | Pairwise distribution | Quantile | TDist | Tweedie

Example dataset | Creating a gbm distribution object | Setting training parameters | Additional variable and fitting parameters | Cross Validation | Parallelisation | Putting it all together | Default behaviour | Identifying the optimal iteration | Fitting additional trees | Predictions on other data | Relative influence of predictors | Plotting the marginal effects of selected variables | Additional useful functions

Set-up of data and distribution object | Creating a boosted model | Strata Updates | Role of additional parameters in GBMDist | ties and prior_node_coeff_var | Description of the underlying algorithm - specifically for CoxPH

High-level description of stochastic gradient boosting | Gradient boosting in more detail | Friedman's gradient boosting machine | Improving boosting methods using control of the learning rate, sub-sampling, and a decomposition for interpretation | Decreasing the learning rate | Variance reduction using subsampling | ANOVA decomposition | Relative influence | Common user options | Loss function | The relationship between shrinkage and number of iterations | Estimating the optimal number of iterations

High-level description of stochastic gradient boosting | Gradient boosting in more detail | Friedman's gradient boosting machine | Improving boosting methods using control of the learning rate, sub-sampling, and a decomposition for interpretation | Decreasing the learning rate | Variance reduction using subsampling | ANOVA decomposition | Relative influence | Common user options | Loss function | The relationship between shrinkage and number of iterations | Estimating the optimal number of iterations

Set-up of data and distribution object | Creating a boosted model | Strata Updates | Role of additional parameters in GBMDist | ties and prior_node_coeff_var | Description of the underlying algorithm - specifically for CoxPH