The correlation between data and an unknown quantity is usually subject to some degree ofeconomic uncertainty. For example, if you knew that 75% of the time a players ' eye twitches when they arebluffing, but such a tic occurs only 10% of the time otherwise, then you would gain some abilityto identify a bluff. But in this case, even if you always spot the tic, you cannot be sure that aBluff is in progress. This is because the model itself has an element of uncertainty built in: thedata has an imperfect correlation with the unknown. In order to make inferences is such casesyou need to build a statistical model to account for the uncertainty, so I will refer to these casesas statistical inference.
đang được dịch, vui lòng đợi..