IQ1: IQ-learning: Recommend stage 1 treatment

In kalinn/iqLearn: Interactive Q-Learning

IQ-learning: Recommend stage 1 treatment

Description

Recommends the IQ-estimated optimal first-stage treatment for a patient with observed stage 1 variables.

Usage

IQ1(mainObj, cmObj, sigObj, dens, h1main, h1cm, h1sig)

Arguments

mainObj	object of type learnIQ1main
cmObj	object of type learnIQ1cm
sigObj	object of type learnIQ1var
dens	method of density estimation, either "norm" for normal location-scale density estimate or "nonpar" for the empiricial density estimator
h1main	vector of observed first-stage main effects corresponding to the variables in H1Main used in learnIQ1main()
h1cm	vector of observed first-stage main effects corresponding to the variables in H1CMean used in learnIQ1cm()
h1sig	vector of observed first-stage main effects corresponding to the variables in H1CVar used in learnIQ1var()

Details

Use the estimated optimal first-stage decision rule from learnIQ1() to recommend the best stage 1 treatment for a patient presenting with history h1. It is essential that h1main include the same variables and ordering as H1Main. If a formula was used to fit learnIQ1main(), we recommend checking summary(<learnIQ1main object>) for the correct order of h2. Similarly for h1cm and h1sig. dens can be chosen by looking at a normal QQ-plot of the standardized residuals from the contrast mean and variance modeling steps.

Value

q1Pos	estimated value of the first-stage Q-function when H1=h1 and A1=1
q1Neg	estimated value of the first-stage Q-function when H1=h1 and A1=-1
q1opt	estimated optimal first-stage treatment for a patient presenting with h1

Author(s)

Kristin A. Linn <[e-mail]>, Eric B. Laber, Leonard A. Stefanski

References

Linn, K. A., Laber, E. B., Stefanski, L. A. (2015) "iqLearn: Interactive Q-Learning in R", Journal of Statistical Software, 64(1), 1–25.

Laber, E. B., Linn, K. A., and Stefanski, L. A. (2014) "Interactive model building for Q-learning", Biometrika, 101(4), 831-847.

See Also

learnIQ1main, learnIQ1cm, learnIQ1var,

Examples

## load in two-stage BMI data
data (bmiData)
bmiData$A1[which (bmiData$A1=="MR")] = 1
bmiData$A1[which (bmiData$A1=="CD")] = -1
bmiData$A2[which (bmiData$A2=="MR")] = 1
bmiData$A2[which (bmiData$A2=="CD")] = -1
bmiData$A1 = as.numeric (bmiData$A1)
bmiData$A2 = as.numeric (bmiData$A2)
s1vars = bmiData[,1:4]
s2vars = bmiData[,c (1, 3, 5)]
a1 = bmiData[,7]
a2 = bmiData[,8]
## define response y to be the negative 12 month change in BMI from
## baseline 
y = -(bmiData[,6] - bmiData[,4])/bmiData[,4]
## second-stage regression
fitIQ2 = learnIQ2 (y ~ gender + parent_BMI + month4_BMI +
  A2*(parent_BMI + month4_BMI), data=bmiData, "A2", c("parent_BMI",
  "month4_BMI"))                                     
summary (fitIQ2)
## model conditional expected value of main effect term
fitIQ1main = learnIQ1main (~ gender + race + parent_BMI + baseline_BMI
  + A1*(gender + parent_BMI), data=bmiData, "A1", c ("gender",
                                "parent_BMI"), fitIQ2)
## model conditional mean of contrast function
fitIQ1cm = learnIQ1cm (~ gender + race + parent_BMI + baseline_BMI +
  A1*(gender + parent_BMI + baseline_BMI), data=bmiData, "A1", c
  ("gender", "parent_BMI", "baseline_BMI"), fitIQ2)
## variance modeling
fitIQ1var = learnIQ1var (~ gender + race + parent_BMI + baseline_BMI +
  A1*(parent_BMI), data=bmiData, "A1", c ("parent_BMI"), "hetero",
  fitIQ1cm)
## new patient
h1 = c (1, 1, 30, 35)
optIQ1 = IQ1 (fitIQ1main, fitIQ1cm, fitIQ1var, "nonpar", h1, h1, h1) 
optIQ1$q1opt

kalinn/iqLearn documentation built on Aug. 4, 2022, 10:15 p.m.