Research: Predicting & Forecasting

[eqcpkkdn73]

Paramnesia: Memory distortion – “Seeing” things in clouds, Jesus on toast, and so on

Forecasting is often stating the obvious, as, “Somewhere next summer will be the hottest day on record,” unless you go on record with a time and location. This is a Prediction. Increasingly, blather and cherry‑picking a poll fills up air time, click bait, as well as, social media.

Predicting is stating XYZ with a criterion for correctness, as well as with a time stamp when made. XYZ will occur this way on this date, for example.

Some Filters to remember:

Establish what you want: Do you want to make money or be a “blabber.” If the former, you must go with the direction the math leads.

[euxunarmz]

Wait as long as possible—BUT NO-O LONGER to still make something possible--to lock‑in” a Prediction, or make a decision. For example, buying or selling a share of Maggot Industries, or voting for Candidate T or C. With “Mad Money,” let someone else take the bullet if Cramer is wrong.

Utilize ANY wiggle room—Obviously, this should not become rationalizing making a decision not to make a decision. Factor in updates for as long as possible, yet still be possible.

[Fezragoreel]

With Dynamic Data, update, update, and update more BEFORE Predicting, because this data is fluid. Avoid believing an election poll until at least a day before the election—UNLESS running a campaign. An unassuming incident can explode. Crap happens out of the blue. For example, the price of oil could shoot through the ceiling if terrorist bomb Cushing (https://en.wikipedia.org/wiki/Cushing,_Oklahoma), after Dallas, Texas, oilman J. R. Ewing, III, makes an appearance on CNBC with a prediction on the price of oil on such and such date.
Polls and survey released to the public become more accurate (or worth noting) nearer an election, because the pollster wants to be the One that was most accurate. To be noted, a new Top Dog emerges, then fades after the system is gamed.

[eyfmeofh22]

Think probabilistically: “Likely,” “highly,” “beautiful,” “ugly,” and so on, tell us nothing, really. Percentages do—often, this will be impossible until close to the event. Certain events can be circled into, because the event has happened. (http://blogs.scientificamerican.com/cross-check/bayes-s-theorem-what-s-the-big-deal/) For example, a missing aircraft. The crash location can be located with mounting evidence of where the plane is not—refining--Utilizing this for as long as possible before deciding. Establishing is‑not and can‑not are both powerful in Predicting.