A less arbitrary method for inferring cause and effect

A method of inferring causality that is based on the concept of signal and noise and does not suffer from being arbitrary is presented. It has been used successfully to determine whether changes in the clinical status of patients with a chronic disease were due to random fluctuations or to the effects of an investigational treatment. The method is not arbitrary because probabilities are compared to the probability of the outcome that is most likely to occur at random (the most noisy outcome) and significance is determined by a unique inflection point on a particular curve.<
>