Description Usage Arguments Value
Cross validation for L1-Penalized Q-Leanring Given group information, the variable selection can be L1-lasso or group-lasso
1 2 3 4 5 6 7 8 9 10 11 |
x: |
n by p matrix of features |
y: |
weights vector of length n |
Yorig: |
original outcome vector of length n |
A: |
treatment vector of length n |
pentype: |
whether penalized Q-learning is used or not, default is "lasso" |
group: |
group number, vector of length (2p+1), have to be consective, in each individual is one group then set group=seq(1: (2p+1)) |
loss: |
default is "ls" for least square loss |
pA: |
propensity score, vector of length n |
nfolds: |
number of cross validation fold, should be an integer >3 |
subject of class "qlearn"
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.