Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
Statistics and probability theory are not easily understood by the general population. These subjects can get very complex. I would suspect that the media would do better of these subjects, but that's just a guesstimate.
But they often fail in the statistical arena. The media might have an additional, perhaps more hidden, reason to distort their interpretation of stats. They have a vested interest in one outcome over an opposing one; they have an agenda. So their analysis is perhaps faulty by ignoring some data sets in favor of selectively using sets that favor their agenda positions. Not honest, not fair, but something that happens often. Too, too bad.
The word “probability” has several definitions. A frequentist definition is "do enough trials and it will happen a determined percentage of the time”. A Bayesian definition is “ it is a measure of uncertainty about an upcoming event". In many instances, doing a meaningful number of experiments to establish a frequentist defined probability is essentially an impossible assignment. So guesstimates rule the day.
Some folks believe that odds favor a change if an event has happened frequently (again and again) in the recent past. For some events that might be a true rule while for others it is a misunderstanding of probability. If a fair coin toss has registered Heads for 9 tosses, what is the probability that it will be Heads on the next toss? Since each toss is independent of the past tosses, the odds are still one-half. Some folks would disagree because of a belief in a reversion-to-the- mean argument. That's bad thinking.
The referenced Nate Silver article is excellent. He doesn't fall into probability traps. As expected, he relies heavily on statistical data and its interpretation. He has consistently used solid statistical methods since his extremely successful "The Signal and the Noise" book. That style of writing worked for that book and works today. Good for him and especially good for his many readers.
A proper use and respect for statistical analyses will make all of us more successful investors. Numeracy is a necessary tool.
Comments
Statistics and probability theory are not easily understood by the general population. These subjects can get very complex. I would suspect that the media would do better of these subjects, but that's just a guesstimate.
But they often fail in the statistical arena. The media might have an additional, perhaps more hidden, reason to distort their interpretation of stats. They have a vested interest in one outcome over an opposing one; they have an agenda. So their analysis is perhaps faulty by ignoring some data sets in favor of selectively using sets that favor their agenda positions. Not honest, not fair, but something that happens often. Too, too bad.
The word “probability” has several definitions. A frequentist definition is "do enough trials and it will happen a determined percentage of the time”. A Bayesian definition is “ it is a measure of uncertainty about an upcoming event". In many instances, doing a meaningful number of experiments to establish a frequentist defined probability is essentially an impossible assignment. So guesstimates rule the day.
Some folks believe that odds favor a change if an event has happened frequently (again and again) in the recent past. For some events that might be a true rule while for others it is a misunderstanding of probability. If a fair coin toss has registered Heads for 9 tosses, what is the probability that it will be Heads on the next toss? Since each toss is independent of the past tosses, the odds are still one-half. Some folks would disagree because of a belief in a reversion-to-the- mean argument. That's bad thinking.
The referenced Nate Silver article is excellent. He doesn't fall into probability traps. As expected, he relies heavily on statistical data and its interpretation. He has consistently used solid statistical methods since his extremely successful "The Signal and the Noise" book. That style of writing worked for that book and works today. Good for him and especially good for his many readers.
A proper use and respect for statistical analyses will make all of us more successful investors. Numeracy is a necessary tool.
Best Wishes