Skip to main content

Table 2 Characteristics of the investigated scenarios

From: Constructing treatment selection rules based on an estimated treatment effect function: different approaches to take stochastic uncertainty into account have a substantial effect on performance

Scenario

Shape of θ(x)

Distribution of X

1

Linear: θ(x)=β(x−0.5)

\(\mathcal {U}(0,1)\)

2

Linear: θ(x)=β(x−0.5)

\(\mathcal {T}(0,1,1/3)\)

3

Concave: θ(x)=β(0.3−0.9(x−1)2)

\(\mathcal {U}(0,1)\)

4

Convex: θ(x)=β(−0.3+0.9x2)

\(\mathcal {U}(0,1)\)