f_report: Report from xgboost predictions

In kristian-bak/kb.modelling: What the Package Does (One Line, Title Case)

Description

Report from xgboost predictions

Usage

f_report(
  ticker,
  target_name = "target_day1",
  from_date = "2016-01-01",
  var_names = NULL,
  buy_cut = 0.6,
  sell_cut = 0.4,
  importance_plot = FALSE,
  shap_plot = FALSE,
  top_n = 7,
  offline = FALSE,
  c = 0.5
)

Arguments

ticker	ticker name
target_name	target name. See options in f_target_names(). Default is target_day1
var_names	predictors used to train xgboost model. Default is NULL, meaning all predictors from f_indicators will be selected.
buy_cut	cut off for when to buy based on predictions from xgboost. Should be between 0 and 1.
sell_cut	cut off for when to sell based on predictions from xgboost. Should be between 0 and 1.
importance_plot	logical. If TRUE, importance plot will be calculated. Default is FALSE.
shap_plot	logical. If TRUE, shap plot will be calculated. Default is FALSE
top_n	top n variables to show in importance plot.
offline	logical. If TRUE, data will be loaded from data folder
c	cutoff for accuracy. Should be between 0 and 1. Default is 0.5
from	date when loading data

kristian-bak/kb.modelling documentation built on Dec. 21, 2021, 7:46 a.m.