Functions & Formulae

Each applied formula has its own function page, with a signature, implementations, and tests.

One-way sensitivity function

f(theta_j,theta0_(-j)) = Y_j

Maps a model output to its value when a single input j is set to a stated value while every other input stays at its base-case value. Running it at the input's lower and upper values gives the endpoints of one bar in a tornado diagram.

  • Swing of a one-way bar

    R_j = abs(Y_jU - Y_jL)

    Takes the absolute difference between the model outputs at the lower and upper values of input j, each run with all other inputs at their base-case values. The swing is the width of the input's bar in a tornado diagram, and bars are ordered with the widest at the top.

  • Switching value by linear interpolation

    theta_jS = theta_jL - INMB_jL * (theta_jU - theta_jL) / (INMB_jU - INMB_jL)

    Finds the value of input j at which incremental net monetary benefit equals zero by interpolating between the results at the input's lower and upper values, with all other inputs at base case. The result is exact when incremental net monetary benefit is linear in the input and an approximation otherwise. NICE calls this value a switching value and asks how far it lies from the current best estimate.