Man pages for timradtke/heuristika
Robust Probabilistic Forecasts to Tinker With
| add_prior_anomaly | Add a prior on the probability of an anomaly |
| add_prior_error | Add a prior on the scale of the error component |
| add_prior_level | Add a prior on the level component |
| add_prior_seasonality | Add a prior on the seasonal component |
| add_prior_trend | Add a prior on the local-linear trend component |
| assert_priors_error_requirements | Assert that either a prior on the error component is properly... |
| autoplot.tulip | Autoplot method for 'tulip' objects |
| autoplot.tulip_paths | Autoplot method for 'tulip_paths' objects |
| default_object | Return an empty 'tulip' object |
| estimate_season | Get the median value for each seasonal period |
| flowers | Flowers time series |
| flowers_holdout | Flowers time series (holdout) |
| fn_get_optimal_param_idx | Get the index of parameters that optimize the log joint... |
| get_fitted | Compute the fitted values given 'level' and 'seasonal'... |
| get_remainder_component | Compute the remaining component in 'x' given that 'component'... |
| get_residuals | Compute residuals from actuals 'x' and fitted values 'fitted' |
| guess_error_sd_from_residuals | Safe initialization of guess for error prior |
| has_large_seasonal_component | Compare residuals to judge whether series has a sizeable... |
| initialize_params_grid | Initialize paramaters to optimize over based on a fixed grid |
| initialize_params_naive | Initialize paramaters to optimize over based on a set of... |
| initialize_params_random | Initialize paramaters to optimize over randomly |
| initialize_season | Initialize the seasonal state given actuals and previous... |
| initialize_states | Initialize state components given time series |
| initialize_states_using_previous_fit | Use a previous fit's initialization as initialization for a... |
| make_default_season_state | Make a default seasonal state vector |
| measure_joint | Heuristic Joint Distribution for Maximum-a-Posteriori Fit |
| plot_components | Plot state components of an 'tulip' model |
| plot_fitted | Plot fitted values of an 'tulip' model |
| plot_forecast | Plot the marginal quantile forecast of a 'tulip' model |
| plot_paths | Plot a few forecast sample paths of a 'tulip' model |
| predict.tulip | Forecast by drawing sample paths from a fitted object |
| repeated_median_line | Fit a level and trend to a time series using Repeated Median... |
| test_prior_error_in_priors | Test the provided prior on error component |
| test_residuals_global_given_y | Test whether provided residuals vector fits to provided time... |
| tulip | Fit a robust exponential smoothing model by... |
| tulip-package | tulip: Robust Probabilistic Forecasts to Tinker With |
