In his book A Mathematician Reads the Newspaper (1995), John Allen Paulos maintains that “much economic and political commentary and forecast are fatuous nonsense”. Because of the inherent “interconnectedness” of the variables of this kind of phenomena, he argues, “we should not expect to predict political or economic developments with any exactitude.”
The author points out that those interconnections can be model through the notion of a nonlinear dynamical system that is highly sensitive to minuscule variations of the initial conditions. These systems “demonstrate a complex unpredictability” and “their trajectories in mathematical space are fractals”. We can rely on social forecasts, Paulos concludes, only if the predictions are short-term, deal with simple phenomena and are just “hazy anticipations rather than precise assertions”.
Most of our personal decisions, however, are based on what we expect to happen in the future --predictions that we usually learn from the media. Moreover, governments, companies, banks and many other organizations use forecasts on hugely complicated political and economic matters to make decisions that often affect the lives of many millions of people.
In practice, “governments rely routinely and heavily on intuitive beliefs about high-stakes outcomes.” [Mellers et al: “Psychological Strategies for Winning a Geopolitical Forecasting Tournament”, 2014] But those apparently intractable problems could be predictable with more accuracy than that assumed by Paulos.
In the recent book Superforecasting: the Art and Science of Prediction (2015), psychologist and researcher P. Tetlock and journalist D. Gardner explain the method used by a small subset of experts who excelled in predicting future events among thousands of participants in a research tournament sponsored by the US government (see Mellers et al).
According to Tetlock and Gardner, “superforecasters” unpack the problem into components; establish a preliminary probability for a given outcome on the basis of how common it is within a broader class –the “base rate” of statisticians--; analyze the specifics of the particular case to adjust the initial probability; explore the similarities and differences between their views and those of others, paying special attention to prediction markets and other methods of extracting wisdom from crowds; express their judgment using a finely grained scale of probability, and update their forecast in light of changing information. Their predictions are like this: “There is a 6% chance that a country will leave the Euro zone by the end of 2016” (see link below).
A different but complementary view is described by L. Smith and N. Stern in “Uncertainty in science and its role in climate policy”, 2011). Based on the economic ideas of F. Knight in Risk uncertainty and profit (1921), they distinguish between “Knightian risk” –future outcomes for which decision-relevant probabilities statements can be provided—and “Knightian uncertainty”, related to outcomes for which probability statements are not possible but are anyway useful to policy makers. We don’t have a clear scientific view of what a 5◦C warmer world would look like –these researchers exemplify--, but science can be certain that the impacts would be huge, even when it cannot quantify them. Communicating this fact may be useful if only because policy makers may erroneously conclude that adapting to the impacts of 5◦C would be straightforward.
Do you think that political and economic phenomena are predictable with some degree of accuracy?
https://www.washingtonpost.com/news/wonk/wp/2016/01/04/how-to-predict-the-future-better-than-everybody-else/
Many different varieties of uncertainty arise in every model. If a model is to used successfully for policy and decision making, the model should consider all uncertainties associated with model predictions.
Information on uncertainty does not make decision making easier, but to ignore it is to ignore reality. Incorporating what is known about the uncertainty into input parameters and variables used in optimization and simulation models can help in quantifying the uncertainty in the resulting model predictions ( the output).
As Niels Bohr expressed it
"Prediction is very difficult, especially if it's about the future"
~ Niels Bohr
http://home.deib.polimi.it/guariso/inglese/Loucks-VanBeek_book/Text/08_chapter08.pdf
Many different varieties of uncertainty arise in every model. If a model is to used successfully for policy and decision making, the model should consider all uncertainties associated with model predictions.
Information on uncertainty does not make decision making easier, but to ignore it is to ignore reality. Incorporating what is known about the uncertainty into input parameters and variables used in optimization and simulation models can help in quantifying the uncertainty in the resulting model predictions ( the output).
As Niels Bohr expressed it
"Prediction is very difficult, especially if it's about the future"
~ Niels Bohr
http://home.deib.polimi.it/guariso/inglese/Loucks-VanBeek_book/Text/08_chapter08.pdf
Philadelphia, PA
Dear Jorge,
This strikes me as a fine question, and it is certainly one worth reflecting on. Partly, I suppose, it is worth reflecting on, because people often presuppose or entertain quite unreflective answers to it. That in itself has significant effects.
It is hard to beat the answer provided above from Bohr, by Ahmadi-Nedushan:
"Prediction is very difficult, especially if it's about the future"
~ Niels Bohr
Its a fine comment. This surely gave everyone a good chuckle!
I would amend the comment from Stoica, though, by Mark Twain, who said:
History never repeats itself exactly, but it rhymes.
People make quite reasonable predictions all the time, and we are often very confident of them: "The sun will rise tomorrow;" "I'll be home by 6pm;" Regarding economic markets, "Whatever goes up must come down;" Or, more general predictions: "People will die tomorrow;" "Children will be born tomorrow," and it is not hard to go on in similar fashion.
What is wanted, though, are more interesting predictions, and part of the problem with these is that any interesting models and predictions on which people come to depend, will subsequently enter into calculations concerning what to do. Any variables, though minor and taken for granted at first, and which support the expected outcomes, may then be exploited to subvert the expected outcomes. On the other hand, constancy of expected outcomes among large groups of people can lead on to "bandwagon effects." For example, toward the end of a big run-up of the stock market, large numbers of "little guys" begin to join in on the expectation of quickly making a lot of money; however, when this happens, the larger investors often take this as a sign to cash in their chips and take their profits--and the upward tendency is then reversed.
In some sense, I think the negative expectations we generally share, concerning the unwelcome effects of nationalisms are quite similar. Its a matter of confident political predictions or warnings. Everyone knows what has happened in the past, and it involves a kind of bandwagon effect. At some point, though, the bandwagon begins to resemble the classical image of the lemmings all running off a cliff together. Introducing diversity helps meliorate the outcome, though this brings its own problems in turn. Where is the point of balance?
People keep interfering to exaggerate or to undercut the most reasonable predictions about the outcomes of human action, and that is part of the reason that the human future remains always up in the air. It depends on what we do, right? "Prediction is very difficult, especially if it's about the future"
H.G. Callaway
Political addition and subtraction is an unpredictable phenomenon particularly when the corruption has overcast the planet.
Dear Jose,
Good question. Economic processes and political events are among the many social structures that are difficult to predict their evolution in time but in periodical manners, society is always hoodwinked by individuals who pretend to know the truth of the future, without even having the capacity to know what such events at present time+epsilon (epsilon- a very small, negligible positive time) will be, what each will be beyond the present. Both structures are chaotic in nature, we have witnessed how the world wars begun, with simple incidents of unnoticed and unpredictable incidents.The human tragedy we see in the middle east are due to unpredictable results in most cases of mistakes of political maneuvering of skewed intent of undermining or not anticipating the consequences of that actions. The economic collapse we witnessed just at no ones knowing, even those economists who secured noble prize did not know what happened and did not know what to do - the difficulty of predictions of evolution of social and economic processes.
In such kinds of unpredictable and chaotic phenomena, it is always necessary to minimize the size of initial conditions we insert in to the system so that huge fluctuations do not happen(we should not perturb things with big), we should not gamble big, not to play with big risks and suffer the unknown consequences. We have to play large, when only we know the outcomes. Predictions of such events are immensely difficult as they both involve human intentions that are guided by thinking (unlike non thinking natural processes) and uncontrolled ambitions.
Dear Colleagues,
Good Day,
"Prediction
A prediction (Latin præ-, "before," and dicere, "to say"), or forecast, is a statement about an uncertain event. It is often, but not always, based upon experience or knowledge. There is no universal agreement about the exact difference between the two terms; different authors and disciplines ascribe different connotations.
Although guaranteed accurate information about the future is in many cases impossible, prediction can be useful to assist in making plans about possible developments; Howard H. Stevenson writes that prediction in business "... is at least two things: Important and hard."
Please, see the first link for detail, if you have enough time see the other references....
https://en.wikipedia.org/wiki/Prediction
http://www.die-gdi.de/en/the-current-column/article/the-ability-of-political-science-to-predict-future-developments/
http://www.theatlantic.com/international/archive/2016/01/global-economy-2016/422475/
http://encyclopedia2.thefreedictionary.com/Scientific+Prediction
http://contemporarycondition.blogspot.com/2011/04/politics-of-event.html
Profetizando acontecimientos es un asunto bien arriesgado. Cuando Fukuyama lo hizo en los '90. augurando el derrumbe de los nacionalismos, los fundamentalismo y el separatismo, el transcurrir de la historia lo derrumbó a Fukuyama. Todos los "ismos" que Fukuyama nos perjuraba que eran muertos-como la historia-se han resucitado con un vigor redoblado. Mejor nos ocupemos de la realidad mientras ocurre, y abstengamos de vaticinios rimbombantes como los que de hecho hicieron mella en el pensamiento académico de los '90.
Dear all! Unfortunately, the history of mankind shows that many of the prospects of development of economy and politics break down in the conditions of the unpredictability of human logic and personalities. Unfortunately, the factor of Homo sapiens is not yet understood.
Dear José Eduardo.
My answer is simple: To the extent that we believe in the model generates the prediction.
I think the beauty of life is that really, although we have the most accurate forecasting model, we can not control (predicting) everything and that's the real magic of life.
Thank you for sharing your interesting question with me. I hope my answer be useful for you.
Regards.
In the field of natural science, weather forecast cannot yet be done to a satisfactory degree of accuracy in spite of the use of big computers and much data from weather satellites. Political and economic processes are much more complicated than weather. So, prediction of these processes would be extremely difficult.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
Wheather forecasts are not accurate, and we are happy since we avoid wars to get the best weather. To predict political and economical processes is more difficult. We have the experience of the past with the previsions of marxism, which were not folllowed by all. Now nobody says that marxism is a science. This bad previsions had bad humam consequences, which must be avoided in the future.
"Politics is not an exact science."
Otto Eduard Leopold von Bismarck. Speech to Prussian upper house (18 December 1863)
"Laws, like sausages, cease to inspire respect in proportion as we know how they are made."
John Godfrey Saxe.
We often do not know the true reasons of many politic and economic processes, and hardly can predict them based on all available information. The exact forecast of the future has a very high commercial and political value and it is usually not in the public domain of discussion.
Dear José,
To predict economic processes and changes needs professional knowledge and experts can have a chance to recognise them. Of course, the reliability of these predictions depends on information, level of professionalism and frankly spoken because of the often too complicated economic processes and the skilfully distributed fake information the prediction may be wrong.
As to political manipulations, the silliness and childishness of voters are determinant. Relatively dim politicians can easily manipulate voters. Regarding Hungarian policy, tricks of politicians were fully transparent and in spite of this they were able to exploit people and destroy their welfare and future. What about in Argentina?
Political & economic phenomena, in the world of today, are largely unpredictable not due to the multitude of variables but because of the lack of the correct or true information by most citizens.
We live in a world in which most politicians lie as much as they breath & we see or listen to media that are well-directed & well-controlled by those who consider themselves as "elite".
I agree with Winston Churchill's saying: "In wartime, truth is so precious that she should always be attended by a bodyguard of lies" but the current political servants extended it to all time.
As for economics, there are those who make a business in forecasts but a wise person knows that economics is much related to politics which means that confusion rules the game that is played on the public for the benefit of the so-called "elite" who possess the "secrets" that give them advantages politically & economically.
I've been reading the predictions for 2016 from some of the leading newspapers and magazines and I'd like to share some with you . I think they may be entertaining and after a while we will see how accurate they were.
Many of the predictions that I've read are dominated, as we would expect, by the forthcoming US presidential election (I must say that forecasters tend to agree that the final race will be between Hillary Clinton and Ted Cruz.and that HC will win.)
Coherently, The Economist seems to suggest that Prediction Markets are more accurate than polls in forecasting election results. Its story about the record of this kind of markets from the late nineteenth century onward is interesting. Financial Times experts seem to take seriously their task to predict a variety of political and economic events, while The Fix (Washington Post) prefers to present its political predictions as a "fun exercise".
http://www.ft.com/intl/cms/s/0/7c70d9a2-aef1-11e5-b955-1a1d298b6250.html
http://www.forbes.com/sites/markrogowsky/2016/01/03/16-undaunted-predictions-for-2016/
http://www.economist.com/news/united-states/21684798-how-jesse-jackson-inadvertently-revived-political-betting-prediction-2016
https://www.washingtonpost.com/news/the-fix/wp/2016/01/01/the-fixs-2016-predictions/
http://www.politico.com/story/2016/01/insiders-predictions-2016-217264
Predictions are not so successful, history is full of such failures.
You have just to read what newspapers 'predict' at the eve of each year and compare it with what happened at the end of the same year: a total disaster!
In predicting a model you have to take into account that a model may not capture the complexities of the underlying phenomena completely. For all intents and purposes what is strived for is "good enough" to work some to most of the time.
One of the underlying assumptions in "big data", machine learning and other trends learning and forecasting is that the underlying distribution that is of interest is stationary or varies so slowly with time that in principle it can be predicted if captured correctly. This works mostly (emphasis on mostly) for natural phenomena.
With respect to human phenomena, that is another matter.This also carries to human activities and social phenomena. One of the major failures of forecasting models in this area is the assumption that players will interact in a rational manner in trying to optimize an utility function (this is especially true for politics). With events and information happening it is dangerous to think that one can capture all the variables necessary for prediction and even if this is possible, it may become obsolete really fast.
Markets are very efficient in absorbing information and therefore the underlying distribution can change really fast. This makes it extremely difficult to forecast and even so it has to be for very short periods of time (you can make some good estimates for cyclicals but even so). It also depends at what scale are you interacting with the market. If you are a market mover, you have enough pull that you can eliminate some of the variables in the system making it easier to predict.
My two cents
Dear José, I went through the references which you provided. Your statement "I think they may be entertaining" is a good one since an entertainment show may include some illusive games & tricks. I saw, in the references, a high dose of "wishful thinking" & some of them seem to be set for current political purposes which means that they called them "predictions" but they are "aspirations" actually.
very well said , dear @ Arturo.
The usefulness of any model depends in part on the accuracy and reliability of its output. Yet, because
- all models are imperfect abstractions of reality, and because
- precise input data are rarely if ever available,
all output values are subject to imprecision. Input data errors and modelling uncertainties are not independent of each other and they can interact in various ways.
Uncertainty in information is inherent in future oriented planning efforts. It stems from inadequate information and incorrect assumptions, as well as from the variability of underlying processes.
https://www.utwente.nl/ctw/wem/education/afstuderen/Loucks_VanBeek/09_chapter09.pdf
Prof. José Walter Pedroza Carneiro, Universidade Estadual de Maringá, Brazil, sent me his answer:
"In my opinion, this type of prediction is impossible. Currently, we are suffering an economic war not economical crisis, and political groups are attempting to destroy our fragile democracy. The economic prediction depends on the professional that is analysing the economic situation. The inflation could be a safety evaluation, but some increases in the prices of goods depends on the greed of every economic player in a situation as we are living. Analysist in television has no credibility. Today, they are using the word crisis, but there is no crisis. In fact, South America is under an economic war. In 2015, Greece disappeared from the Television. How to believe in data collected by economists at right or at left? We have to analise the past by our own and make the right decision who will be the President in the future. South America has paying the development of USA and Europe along the time. That is true and we can believe. Pay attention on the couple of Nobel Prize that open a company and rapidly went to the bankruptcy because the time lag of the analysis was small enough to predict their results."
I think there are important social, political and economic phenomena that can be predicted within certain limits .An example, in my opinion, is the set of processes triggered by the industrialization of society --what some have called "modernization" , or just "development" .
Britain was the first society that experienced this process in the eighteenth century, and since then many other nations -- by now almost all - have followed the same pattern, although at different rates and levels of development.
This sequence is so regular that can be predicted with accuracy. Once any society takes the path of industrialization, we know that people will migrate from the countryside to the cities. Therefore, urban population will grow. Gradually, we will see an increase in the division of labor, life expectancy, literacy and the dissemination of scientific knowledge. Simultaneously, family size will be reduced and women will enter the market labor. We can also predict that political participation will grow (remember that liberal democracy is not the only system that promotes participation). There are other phenomena that accompany this process and I do not mention to synthesize.
Furthermore, we can highlight two additional points :
1) Many societies have gone through this process even though at different times they have experienced setbacks, massive destruction due to war, etc. ie , they have often gone by very high levels of chaos ;
2) Contrary to what some authors state, these phenomena are predictable in the long run, over the chaos and uncertainty related to short term process.
I do not see much difference between this type of prediction and that made in many areas of natural sciences. Astrophysics, for example, can predict the destruction of the Earth when the Sun, running out of fuel, will expand, engulf and incinerate our planet. It cannot predict, however, many details in the meanwhile. For example, an asteroid can hit the Earth and destroy most life forms, etc.
Jose,
Thanks for the question it is a very interesting subject.
You are right, there are some phenomena that exhibit some kind of predictability over longer intervals but follow stochastic behavior over shorter periods of time (and the converse is also true of other phenomena). My particular cautionary stance is that: broadly one cannot broadly state that we can predict phenomena under politics nor economic events.
With respect to your example of modernization, I would debate the statement, if we narrow it down by:
1) Sector. Some sectors will stagnate while others move with unprecedented speed
2) Geographical region(some regions are just consumers and not significant producers)
3) Specific goal of modernization
This where it gets interesting but at the same time difficult to predict.
Another point is that eventual movement of industrialization will level off because of the basic principle of scarce resources. While up till now we have postponed reaching the limits of our resources by creative optimization, I see this as leveling at some point (we now have your prediction and mine !-)
Dear Jose/All,
"Meet 2016's Worst economic Performers":
Venezuela-3.3%
Brazil
Greece
Russia
Ecuador
http://www.bloomberg.com/news/articles/2016-01-11/meet-2016-s-worst-economic-performers-flirting-with-disaster
Dear Jose,
We know that Meteorologists use data collected painstakingly and analyze it through models that are mathematically unchallenged. Yet when they say it will rain in the next 24 hours, we end up having bright sun shine, If pure scientific data processed through mathematical models can fail so, believing that political forecasting could be accurate is just plain wish-making.
We just need to look at the phenomena of elections, be it municipal, regional or national. The political pundits and media use a variety of surveys and models to forecast outcomes and at the end ordinary citizens manage to spring surprises. Even where muscle power and corrupt ways are rampant to subvert the political process, outcome of the election could never be predicted correctly. Take the general elections in India a year back. Everyone felt that anti-incumbency factor and corrupt image of the ruling party then will run them down. But no one ever thought that the winning party will accrue such thumping majority. Some of the oldest ministers in the erstwhile party had remained elected from their constituency for three decades. The question of them loosing out was unimaginable, even though their party was predicted to loose. Yet they lost, and lost by huge margins.
Economic forecasting is one shade better because the number and variety of statistically manageable data is large. But this forecasting has two major pitfalls. The first is explained by the simple quip - lies, damn lies and statistics. My apologies for being blunt. But, these forecasts depend on who is asking for the forecast and for what purpose. When the government wants to show that they have lifted millions out of poverty, they need to just lower the bar to show that "x" million poor are now above the bench mark. When a corporate bids for certain concessions from the government, they need to statistically show that their proposal is environmentally friendly, pro-employment and economic development of the local community where the business is to be established.
The second pitfall is - when we say the country's GDP has grown by "Y" points or industrial production can increase by "Z" per cent by adopting to reforms, do we actually mean "X" number of poverty index will come down? Yes, economic forecasting is a useful tool to predict the outcomes through manipulation of given data through models, and it is certainly a handy tool for shaping policy. Whether such modeling actually changes the destiny of the poor and underprivileged whose numbers seem to increase by the day, is a question.
In nut shell, accuracy in both the spheres is just a matter of perception. Statistically, yes; realistically, not so.
Dear Arturo:
Yes, you're right. As we ask more and more specific questions , it also becomes more difficult to predict phenomena. But this may just mean that we need more data or work more on theory development In the limit, probably no phenomenon in any field can be predicted in each of its details.
Political and economic events can be predicted as far as probability theory can determine the position of the electron.
Economists are unable to predict Disruptive events
"Tetlock systematically collected a vast number of individual forecasts about political and economic events, made by recognized experts over a period of more than 20 years.
He showed that these forecasts were not very much better than making predictions by chance, and also that experts performed only slightly better than the average person who was casually informed about the subject in hand. Depressingly, he found that there were remarkably small gains to greater expertise when it came to forecasting accuracy.
This result, which applied to politics and international relations as well as economics, emphasizes how hard it is to forecast disruptive or discontinuous events – in fact, events which are not indicated by relatively simple extrapolations of recent past trends.
The credit crisis was certainly not one of those “forecastable” events. If we ask why economists failed to predict the credit crisis, we should also ask why political scientists failed to predict the recent Egyptian rebellion, or indeed a terrorist event like 9/11. Or why seismologists cannot predict earthquakes...."
http://blogs.ft.com/gavyndavies/2011/02/10/why-cant-economists-predict-disruptive-events/
Dear Behrouz:
You reproduce the following:
"He (Tetlock) showed that these forecasts were not very much better than making predictions by chance,"
This statement refers to the average result taken from all forecasters. The interesting study --which I cited-- refers to the small set of superforecasters, who systematically perform much better than chance.
As a side note, biologists and physicians can not usually predict mortal epidemic diseases. This does not mean that their disciplines are unable to make good predictions.
I think one has to be cautious, because completely deny any form of predictability in economics and politics would mean leaving those fields outside the domain of science.
Wouldn't this mean that there are areas of reality that are beyond the scope of science?
Dear José,
You wrote:
"I think one has to be cautious, because completely deny the predictability in economics and politics would mean leaving those fields outside the domain of science"
The conclusion of the Tetlocks study are related to prediction of disruptive events only and can not be generalized to all predictive models in economics and politics. It is easier to build model for normal condition.
I suppose that predictive models in economic and politics may involve more uncertainty than the models in other fields of science due to human interaction, unforseen disruptive events, etc.
Best regards,
Behrouz
Dear Behrouz:
I agree. With my call for caution I was not referring to your contribution, but speaking more generally.
Jose,
What In my opinion will limit the predictive power is the complexity of the environment being forecasted and the time frame used.both of these elements are present in politics and economics. In this types of scenarios the small errors that happen in the model tend to accumulate as we extend the time period and eventually make accurate forecast loose their value. If this error accumulation did not happen, then most likely the variable is not significant to the underlying processes (it might be argued that some later decision overtook the decision process and "corrected" the process, but this argument would support that the initial model was worthless from the beginning).
Is prediction more art or science?
"Depends on the domain.
In poker—or, more generally, casinos—it is closer to a science, where classic frequentist statistics apply to repeatable games with well-defined sampling universes: There are only so many cards in a deck.
But in the messy real world, it is much more of a mix of art and science. Carefully crafted questions and good feedback help move the dial toward science, but subjective probability estimation necessarily requires an artist’s touch...
...good forecasters follow processes, but not rules..."
https://www.reit.com/news/reit-magazine/march-april-2016/phil-tetlock-explains-art-and-science-superforecasting
The problem of the "human factor" is the key to development and destruction of entire generations and communities. This problem is irreversibly fatal in the field of aviation and space. Please refer to the documents and articles in the attachment.
Macroeconomic forecasting is a very difficult task due to the lack of an accurate, convincing model of the economy. The most accurate models for economic forecasting, “black box” time series models, assume little about the structure of the economy. Constructing reliable time series models is challenging due to short data series, high noise levels, nonstationarities, and nonlinear effects.
Macroeconomic modeling and forecasting is challenging for several reasons:
No a priori Models: A convincing and accurate scientific model of business cycle dynamics is not yet available due to the complexities of the economic system, the impossibility of doing controlled experiments on the economy, and non-quantifiable
factors such as mass psychology and sociology that influence economic activity.
Noise: Macroeconomic time series are intrinsically very noisy and generally have poor signal to noise ratios
Nonstationarity: Due to the evolution of the world’s economies over time, macroeconomic series are intrinsically nonstationary.To confound matters, the definitions of many macroeconomic series are changed periodically as are the techniques employed in measuring them.
Source:
Moody, John, "Economic forecasting : challenges and neural network solutions" (1995). CSETech. Paper 266. http://digitalcommons.ohsu.edu/csetech/266
http://digitalcommons.ohsu.edu/cgi/viewcontent.cgi?article=1265&context=csetech
Dear Colleagues,
Good Day,
"The best way to predict the future is to study the past, or prognosticate."
----- Robert Kiyosaki
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Probability