Talk:Prediction

can you please help me know my future

Deleting or fixing?

Wow, this article is crap. I can't see it ever containing anything useful. Anyone mind if I nominate it for deletion? --Doradus 02:53, Dec 15, 2004 (UTC)

It would be well suited for an entry on the Scientific Method, prediction is a bit of a stretch and perhaps a bit misleading. --johnycanal 03:58, Feb 15, 2005 (UTC)

I don't agree. I'm trying to fix the article, but it will take a lot of work. In particular, I don't like the "future topics" section. Much of that should be integrated into the article. --Pablo D. Flores 14:17, 27 Apr 2005 (UTC)

It could use fixing, but come on, who actually has the patience to read "predictions"? -J.H.

Please also consider adding some external links: http://www.longbets.org http://www.zapfuture.com http://www.icalledit.com/forums

Cleaning up the mess (list of topics)

I'm moving these sections here. The main issues should be integrated into a few short well-wikified paragraphs, maybe following a timeline. {{mainarticle}} should be used for the larger sections. Whatever doesn't fit in running text should be put into a "See also" or "Related" made entirely of relevant links, and containing no more than 8-10 entries if possible. --Pablo D. Flores 14:38, 27 Apr 2005 (UTC)

Future topics

• past approaches to prediction
• the mechanistic universe
• the omniscience of God
• chance and probability
• modern scientific ideas
• is it possible to predict the future, even in principle?
  • quantum theory
  • limits on computability, and the holographic principle
  • chaos theory
  • phase space evolution
  • relativity theory and "block time"
• Prediction and forecasting in practice
  • Statistics
  • Simulation
  • Weather forecasting
  • Economic forecasting
  • Transport planning
  • Stock markets and Futures markets
  • Think tanks and futurology
  • Management of risk, for example, the Insurance industry
  • Extreme-value theory

'Predictions' is a very broad topic. For it to be a credible page, it can be divided better. If someone can use a mind-map to make some sort of outline that would make this page valuable to everyone. For example: Credibility -> Verifiability VS Falsifiability Links Links, etc.. Educated Guess VS Intangible Unprovable Evidence (Even if future) and possibility VS (Pessimism Vs Optimism vs Pygmalion Effect) VS Useful Construct etc.. Also, you can have titles such as 'Predictions In Business' or 'Predictions In Everyday Life(Common Sense, Estimation)' or 'Predictions In Superstition/Insanity/Depression & Pessimism' (Omen, possibly Mechanistic Universe, possibly Ying/Yang) etc.. 218.186.8.12 19:59, 11 December 2006 (UTC)P. J. Claudio

Related topics

• talk:Scientific predictions - Ideas for expanding the Prediction article
• failed predictions
• falsifiability
• fortune-telling
• free will and consciousness
• future of...
• gambling
• hidden-variable theory
• hypothesis testing
• inference
• linear prediction
• logical induction
• oracles
• prediction interval
• probability theory
• prophecy
• psychohistory
• quantum indeterminacy
• randomness
• receiver operator characteristic
• retrodiction
• scientific method
• Seymour Geisser
• statistical prediction

What about statistical forecasting?

I think it is a big mistake to redirect "Forecasting" to "Prediction". I was hoping to find an article about statistical forecasting which would have inclduded details about many time-series quantitative statistical techniques including exponential-smoothing, ARMA, and the many many more. It was a big disapointment to be re-directed to this article - it is like re-directing enquires about "Astronomy" to an article about "Astrology"!

I have written some more comments about this here - http://en.wikipedia.org/wiki/Category_talk:Statistics

It is, by the way, more correct to refer to "statistical forecasting" than "statistical prediction" since all the textbooks about this topic that I have seen use the word forecasting in the title rather than the word prediction.

Thanks

fred2

I have applied your suggestions in Forecasting. Also see Predictive analytics

Apdevries 15:19, 2 September 2006 (UTC)

Unnecessary and distracting page

I agree with the previous correspondent: why is there a page on "prediction" as well as one on "forecasting". This page should be removed or, failing that, links should go from prediction to forecasting but not the other way around. --Kesten 01:07, 16 November 2006 (UTC)

A forecast generates a prediction. If the prediction is accurate, the forecast was correct, and this information adds weight to the validity of the forecasting technique for planning purposes.

Prediction is just a guess about the future; forecasting, however, uses a more scientific method to arrive at its results. Think outside the box 11:26, 2 March 2007 (UTC)

The term prediction is ok in a religious context. In science the relevant term is forecast. Everything on science should be removed from this article. Gabriel Kielland 16:23, 2 March 2007 (UTC)

Disagree. Forecasts are a branch of the whole concept of predictions. Only, they're made with the help of models, computers or scientific data. But it remains that time is the only way to learn whether they are right or wrong. --Childhood's End 16:54, 2 March 2007 (UTC)

The future as a chaotic system

If future events may be described as chaos,the chaotic nature of the system may be a basis for a future edit to this article. Since chaos is unpredictable,we cannot predict what events will occur,so we narrow it down to a finite set of events that are plausible. For example,it is plausible that earthquakes will occur near known fault lines,since they have occured there in the past. It's not plausible that flying saucers will land in Nashville to meet with Elvis. Prediction is merely a process of pointing out a future event that is so plausible,it seems likely to occur. For example,suppose Barack Obama gets most of the popular votes in the November election,and he appears to win,but Mc Cain gets enough electoral votes to put himself into the White House. I predict that millions of Blacks all over the country will become angry,and start riots,to protest,because they will feel that Obama had been cheated out of a legitimate political victory. Predicting a riot,like that,is plausible,because riots have occured,for political reasons,in the past. If you read Wikipedia's article on Chaos,it says chaotic systems look unpredictable,but they aren't,they behave according to natural laws. Anthony Ratkov —Preceding unsigned comment added by 69.218.77.191 (talk) 06:18, 13 October 2008 (UTC)

Chaotic systems are not unpredictable -- they are unpredictable over long horizons. Weather is a good example -- predictions out to a few days in the future are extremely accurate. Predictions six months in the future are meaningless. Wikiant (talk) 13:00, 13 October 2008 (UTC)

Article about significant long term predictions in history

Does an article exist that strictly covers accurate predictions from the distant past that turned out to be true, or one that summarizes the accuracy of various historical futurists? I've been able to find isolated examples, like Newton's prediction of the existence of Neptune, but I remember in school we talked about books people had written from time to time in history that predicted the future of science and society, some more accurate than others - there seems to be a lot of stuff about present day futurists, but what about historical ones? Where should I be looking? Thanks 131.107.0.105 (talk) 17:49, 11 August 2009 (UTC)

what is my feature now,after 10 years? —Preceding unsigned comment added by 112.201.94.245 (talk) 04:39, 23 January 2010 (UTC)

Practical Relationships for Predicting the Future

We predict the future from moment to moment with high probability of success; so long as our reference frame is a few minutes. Examples: Our intent to go to our car and start the engine, to slow down because of an anticipated bump...

The further out we can make predictions, the greater opportunities are presented to us. A laborer's insights yield them lower pay and Project Managers are paid much higher wages to make predictions on a regular basis; to ensure that a prediction (project) comes to pass. The Space Shuttle was a prediction and great effort was taken to ensure it came to be. Each of the efforts were predictions that were actively acted upon.

Whether or not a prediction occurs because of passive or active intent, does not matter. The prediction was made, and the outcome confirmed (self fulfilling prophecies).

Escalating Opportunities occur when predictions can be made on a consistent basis. Anarchy does not allow for long term predictions because there are no rules. Ethical relationships provides frequent creation of new opportunities because of predictable outcomes.

The following chart provides commonly known relationships, that allow a person to relate thier feelings to predictable outcomes. If they want to have opportunities offered to them, then they need to understand their share of the related responsibilities; mutually being able to predict their future.

File:Predicting the Future - Knowing Outcomes Creates Opportunities.jpg

Add caption here

The chart corelates many factors in such a way that a person can relate their own experiences to the chart and through Probability Aliasing (reading between the lines), they can make better choices for themselves. — Preceding unsigned comment added by Jamesbdunn (talk • contribs) 10:04, 30 December 2011 (UTC)

Predicting across space rather than time

I wonder why there is no attention paid to making predictions across space rather than time, such as predicting patterns in data? Isn't it all the same concept?

The forecasting page said 'Prediction' is the more general concept, so I don't think it would be valid to counter that we could just lump spatial prediction under forecasting instead. LegendLength (talk) 03:19, 27 May 2013 (UTC)

Undue Weight

Like CAPM, there is only one model that fits this niche, certainly no other predictive project-decision models of this sort: Non-regression, nonstochastic, etc..GESICC (talk) 22:33, 8 August 2016 (UTC)

A couple of questions:

1. Could you elaborate on how RiskAoA relates to CAPM?
2. Can you provide a reliable secondary source that supports your claim that there are no other predictive project-decision models?

I am not an expert in this area, but I know that there are many decision support systems including information aggregation tools, qualitative and quantitative scenario planning tools, multiple-criteria decision analysis methods, and so on. I have not found any reliable secondary sources that indicate that RiskAoA is preeminent in this area. Biogeographist (talk) 01:14, 9 August 2016 (UTC)

A They are both unique to their niche, one tool, one function. Intuitively, CAPM is Strategic, RiskAoA is operational, while PRA is tactical.

B How can I show something that doesn't exist? Your decision support system link expressly defines how RiskAoA is different. It doesn't use those techniques. V/RGESICC (talk) 04:50, 9 August 2016 (UTC)

Thanks for the response, but it would be helpful if you would spell out the acronyms you use or wikilink each acronym to the relevant article. Remember that Wikipedia is edited by a very diverse set of people and everyone may not interpret your abbreviations as you intend. For example, the first thing that comes to my mind when I hear the acronym PRA is participatory rural appraisal but you are more likely referring to probabilistic risk assessment. I find your following statement to be unclear: "Your decision support system link expressly defines how RiskAoA is different. It doesn't use those techniques." Which techniques are you referring to when you say "those techniques"? On the contrary, it seems obvious to me that RiskAoA is a (Microsoft Excel-based) decision support system. As the article on decision support systems states, "Some authors have extended the definition of DSS to include any system that might support decision making." Biogeographist (talk) 11:38, 9 August 2016 (UTC)

Also, I wasn't asking you to "show something that doesn't exist"; I was only asking for a reliable source that corroborates your claim that "certainly no other predictive project-decision models of this sort" exist. The RiskAoA steps currently listed at Decision-making § Steps (which are unsupported by a reliable source, but let's assume for the purpose of the present discussion that they are accurate) sound like a typical multi-criteria decision analysis procedure. Therefore there are many other similar methods; see, for example, Multiple-criteria decision analysis § MCDM methods. Biogeographist (talk) 17:26, 9 August 2016 (UTC)

Acronyms, no objection, no topic of discussion, either. Part two of your argument, being Excel based, is non-germane, tools used to be paper-based, the underlying math is what is unique. Also, a dowsing tool is a novel decision tool, you are zig-zagging between objecting to general and specific traits using examples from both that do not apply. This also has nothing to do with undue weight.

You have not made a germane argument in paragraph 2, if you believe that the examples you give provide rival approaches, then by all means, list them. GESICC (talk) 20:42, 9 August 2016 (UTC)

Microsoft Excel appeared in a parenthetical remark; it was not part of my argument, so yes, it was a "non-germane" detail (but a technologically interesting detail, in my opinion, given that the article on decision support systems states that: "Holsapple and Whinston classify DSS into the following six frameworks: text-oriented DSS, database-oriented DSS, spreadsheet-oriented DSS, solver-oriented DSS, rule-oriented DSS, and compound DSS"). My second paragraph, however, was very much germane. I would not call the list of MCDM methods "rival" approaches, but they do illustrate the variety of methods available. I am not interested enough to spend time adding to Prediction § Project risk management myself, but even a casual search on Google Scholar convinces me that RiskAoA is not the only solution to the problem of project risk management. Biogeographist (talk) 21:12, 9 August 2016 (UTC)

Thank you for your consideration.GESICC (talk) 21:29, 9 August 2016 (UTC)

As prediction is perhaps the single most important process in all of nature, this article underserves in the extreme.

Prediction in any quantity, so long as it outperforms raw search, is evolutionarily advantageous, is analogous to "fitness". Its what determines relative survival in any competitive situation. Prediction is synonymous with intelligence and evolutionary fitness, with compression and information, with thermodynamics and entropy maximization, with graph theory, with topology theory, with network theory, with locality, with least action, etc. This article seems to be written by someone who is entirely unqualified to contribute to or report on or document or condense the topic. I am not sure how wikipedia is structured as an editorial and content creation service, but there appears to be a fatal disconnect between ephemeral facts out on the terminal branches of the knowledge tree and the capacity of Wikipedia to provide foundational articles describing the main foundational branches of knowledge. Prediction is one such topic, foundational, keystone, a main-branch concept deserving of an article that does service to a topic of this weight.

Proposed merge of Predictive modelling into Prediction

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

To not merge; topics are sufficiently distinct to warrant separate coverage; opposition and no support. Klbrain (talk) 08:22, 14 October 2023 (UTC)

---

More specifically, section Prediction#Statistics; as per definition: "Predictive modelling uses statistics to prediction outcomes". fgnievinski (talk) 01:09, 4 April 2023 (UTC) fgnievinski (talk) 01:09, 4 April 2023 (UTC)

DON'T MERGE - My bad knee might predict that it will rain tonight. My mother-in-law predicts a trillion things. Predictive modelling is fundamentally (ideally) a scientific endeavour. We also don't equate politics with the "scientific" field of political science. Bquast (talk) 15:17, 22 June 2023 (UTC)

The definition is, the definition is something that is new 2601:240:E083:6D00:2CCE:1E6F:A3B3:D0F0 (talk) 01:11, 21 September 2023 (UTC)

No 2601:240:E083:6D00:2CCE:1E6F:A3B3:D0F0 (talk) 01:12, 21 September 2023 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.