dml_boosting {DMLLZU} | R Documentation |
Double Machine Learning based on boosting
Description
The biggest difference with other method, the trees of this method are generated sequentially. Each tree is constructed using the information of the previous generated trees. Each tree is generated according to a modified version of the original data set, and finally these trees are combined to establish a prediction model
Usage
dml_boosting(y,x,d,data,sed)
Arguments
y , x , d , data , sed |
Value
y Dependent variable;
d Independent variable;
x Control variable;
sed A random seed;
data Data
Author(s)
Lixiong Yang<ylx@lzu.edu.cn>
References
Jui-Chung Yang,,Hui-Ching Chuang & Chung-Ming Kuan.(2020).Double machine learning with gradient boosting and its application to the Big N audit quality effect. Journal of Econometrics(1),. doi:10.1016/j.jeconom.2020.01.018 Victor Chernozhukov,,Denis Chetverikov,,Mert Demirer,... & James Robins.(2018).Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal(1),. doi:10.3386/w23564.
See Also
https://github.com/lixiongyang
Examples
library(ISLR)
attach(Auto)
data<- Auto
y <- data$mpg #Dependent variable
d <- data$origin #Independent variable
x="weight+year +horsepower" #Control variables;
dml_boosting(y,x,d,data,sed=123)