Interactive explainer

Interactive: Value of information: when is a test worthwhile?

Information has value when it can improve a decision. EVPI values perfect information, while EVSI values a particular imperfect test; subtracting test costs gives its expected net benefit.

Interactive

Interactive: Value of information: when is a test worthwhile?

Change the demand probability, test accuracy and cost. The complete decision tree is calculated locally in your browser.

Reading the display

The three bars compare expected value without testing, with testing after cost, and with perfect information before cost. The accompanying text reports EVPI, EVSI and net benefit. A negative test value means costs outweigh its expected advantages. The table shows each signal’s probability, the updated demand probability and the action selected afterwards.

Launching earns €80,000 under high demand and loses €40,000 under low demand. Declining means zero. Keeping this payoff table fixed reveals how knowledge alone changes the decision. The example does not value any real company and requires no personal information.

Three experiments

  1. A useless test: set accuracy to 50%. EVSI becomes zero; increasing cost makes testing correspondingly worse.
  2. Perfect information: set accuracy to 100% and cost to zero. The test reaches perfect information, and EVSI equals EVPI.
  3. The action threshold: change the probability of high demand. Watch the table to see when a positive or negative signal still changes the action.

Reset restores a 40% high-demand probability, 80% test accuracy and €5,000 cost. Sensitivity and specificity are equal in this example. Delays, changing demand and additional actions are not included. The full value-of-information article explains the calculation and its limitations.

From NyxAI

smartlytics is the learning hub of NyxAI GmbH.

Explore the methods we work with, explained openly, with sources and calculations you can reproduce.

Decision Core

NyxAI is developing a Decision Core for Decision Intelligence that combines optimisation, simulation, statistics and reinforcement learning: compare options, make uncertainty visible and learn from outcomes, with people in control.

AI workflows at NyxAI

DeepSquare

Explainable chess AI with human playing styles: NyxAI’s testing ground for understandable decisions under time pressure.