When Risks Compete, a Curve Can Help Predict Which Event Is More Likely to Occur
An approach developed by Claudio Del Sole, Antonio Lijoi and Igor Prünster makes it possible to study how long a phenomenon lasts and what causes it to end. Central to their method is the new concept of “prediction curve”, a tool that shows which event is most likely, depending on when it occurs. This idea can make complex survival-analysis problems easier to interpret and may prove useful in fields ranging from medicine to system reliability