Ülo Ennuste majandusartiklid

“RISK SAVVY” ehk kuidas teha häid riskiotsustusi

Kalkuni illusiooni valemist ning  vaimulik Bayes’i mudelist jne – Gerd Giegerenzer (2014) „Risk Savvy“* ALLEN LANE 322 p

Nt lk 37:

„Tõenäosus et midagi kordub nagu see on eelnevalt toimunud n korda = (n+1)/(n+2)“

See on Pierre-Simon Laplace (1749-1827) nn  “järjestikkuse reegel” – nt kui kalkun on saanud peremehelt korra süüa siis ta arvab et tõenäosus selle kordumiseks on 2/3 aga järgmine päev juba 3/4 (seega suurem) jne kuni peaaegu kindla uskumuseni: lapsus on selles et kalkun ei tea midagi Tänupühadest mida peremees rangelt peab ja seega kurvalt lõpetab.Muide kui kalkunil olekski mingit intuitiivset/subjektiivset aimu Tänupühadest siis nii primitiivses mudelis seda ei saaks formaliseeritult käsitleda.

Lapuse üheks makro-küber näiteks parajasti võib tuua nt meie võimurite väited et „suurt narratiivi pole vaja“ ja „olulisi maksusüsteemi muudatusi ees ei seisa“ jne – lapsus sest nt EL Komisjon on juba üle tosina aasta (n>12) talunud meie täiesti  kodumaiselt ebaõiglast ning rahvusvaheliselt kõlvatut 0-kasumiaksu süsteemi – meie võimurid seejuures moraaliriskiliselt teadmatuses/varjates/tabuteemastades –  et EL Parlamendis ning Komisjonis (meie esindajate osavõtul) veel tänavu jõustatakse  reformid** et kogu EL28 ulatuses maksusüsteemide ebaõiglused ning kõlvatused likvideeritakse – mis meie jaoks võib kurvalt lõppeda (nt eurotoetuste oluliste kärbetega sest nt viimatistel aastatel on Eestist häbiväärselt välja sahkerdatud rohkem investeeringuid kui Brüsselist on toetusi meile heldelt transfeeritud).

Kahjuks meie võimurite makroküberneetilisi suurimaid teadmusvaeguseid näib olevat olevat täielik populistlik mõistmatus rahvusliku jätkusuutlikkuse kestlikkuse riskiarvestuste alal (hiljutine kahetsusväärne näide nt et ka nn      e-residentsuse mehhanismi jõustamisel jäid riskiarvestused tegemata nt isegi võimalike kriminaalse orientatsiooniga rahvusvaheliste agentide võimalike tegevustega seotud kaasuste osas – mehhanismi taolise disaini puhul)

Nt lk 265 – käsitletud Praost Thomas Bayes’i (1702-61) reeglit milles on võimalik subjektiivset tõenäosust formaliseeritult modelleerida. Olla aga see häda et paljudel olevat raskusi mõistmisega sellel formalisatsiooni kujul (G.G. väitel ja ka minu kogemustel tõepoolest) nagu ka üldse tõenäosuse intuitiivse mõistmisega (reegel üsna detailse matemaatilise mudelina sellel leheküljel antud tinglike tõenäosuste alusel) – kuid raamatus on sellest dilemmast väljapääsuks antud tõepoolest rida adekvaatseid lihtsustusi Bayes’i originaalmudelile: eriti nt lk 261 nn „naturaalsete sageduste“ kontekstis (muide lihtsustuste versioone vt nt ka Wikipedia ja märkimisväärselt lihtsustatult nt Collins English Dictionary 2000 lk 133 : kus tinglik tõenäsus P(E/A)=P(A/E).P(E)/P(A)) ning eriti P(A/E) võiks olla tõlgendatud kui “subjektiivne/intuitiivne tõenäosus” – muide – vist peaaegu ainuke valem selles ligemale 1800 lk magnumis)

Tõepoolest väärt lihtsustused nt: intuitiivselt on väga raske ikka mõista nt Oxfordi mõningate professorite 2015 prognoose (FT.com 13.II 15 Twelve ways the world could end) nagu oleks parajasti Tsivilisatsiooni TuumaViimsepäeva tõenäosus 0,00005 – või midagi taolist kusjuures langenud võrreldes poole sajandi tagusega (selles kontekstis  eriti huvitav on nn võimalike „teadmatute teadmatuste“ võimalike riskide arvatamine:

„ …Unknown consequences”: “A catch-all category to cover the unknown unknowns — an amalgamation of all the risks that we have not thought of or that seem ridiculously unlikely in isolation (such as sending signals to extraterrestrial civilisations that attract deadly alien attention). Together they represent a significant apocalyptic threat. Probability: 0.1%“ (sic!üe))

Märkimisväärne et Professorid on  selles lõigus ilmselt uudselt kombineerinud nii Taleb’i Musta Luige ja Rumsfeld’i tundmatute sündmuste ning Zadeh’i hägusloogika kontseptsioone – oleks ka raamatust soovinud leida laiemalt taolisi käsitlusi

Samas nt ka The Economist 7.III 2015 lk17 „Briefing Nuklear weapons – The unkicked addiction“ annab illustratiivse TuumaViimsepäeva saabumise kella näidud (osundus Bulletin of the Atomic Scientists)– mis sümboolselt minutites/tiksudes näitavad et kuipalju on veel aega kosmilistes minutites jäänud Öö saabumiseni (kui midagi otsustavat globaalselt ette ei võeta kellaaja tagasilükkamiseks – vt ka H. Kissinger 2014 „World Order“ lk331 pt „World Order in the Nuclear Age“ – tõsi küll 2015 tuleks seoses putinoididega rääkida „NEW Nuclear Age“) – neid minuteid tiksuda olla praeguseks aga vähem jäänud kui pool sajandit tagasi – seega ratsionaalsete riskiarvestustega otsustusreeglite kohaselt on ilmtingimata kibekiiresti tarvis rahvusvahelisi tuumarelvade täieliku likvideerimise lepinguid – kas või sunniviisiliselt kui efektiivsest sanktsioneerimisest ei piisa***.

Moraal: raamat väärib täiega oma alapealkirja “How to make good decisions” –  meil eriti sest eliidi sügava makroküberneetilise vaegteadmuse tõttu isegi võimalike rahvuslike moraaliriskiliste/korruptsiooniliste kaasuste käsitlused jäävad makro-otsustustest välja: nt nagu mainitud e-residentsuse (mehhanismi just selle disaini) jõustamisel, aga ka majandusliku ebavõrdsuse süvendamisel, kapitali väljaveo mahitamisel, sinise tunneli ja euroraudtee disainimisel, piirirajatiste rajamisel, sanktsioneeritavate firmade subsideerimisel 0-kasumimaksu mehhanismi varjus, Soomlaste ja Nobelistide halvustamisel, jne jne.

Muide ühes G.G. väites julgen kaasa rääkida nagu et riskide arvestamisel ei jää midagi ratsionaalset teostamata: julgeks lisada et riskide arvestamisel toimuks palju ratsionaalsem ja ka mahuka küber-makro tegevus eriti positiivsete riskide teostamisel – nt Eestis oleks piirirajatised ammu korras nagu Soomes, meie kodumaised investeeringud oleks kordades kõrgemad pc nagu Soomes (ja ei asetseks suuresti mitte Jordaanias või Küprosel), majanduslik ebavõrdusus ei oleks barbaarne ja on ohtlik taolise etnilise heterogeensuse puhul arvestades Ida-Piiri tagust määramatust, rahvusliku teadmusruumi kvaliteet ei kõduneks (nt teadusministriks oleks teaduskraadiga rahvusvaheliselt tunnustet persoon, valimiskampaaniates ei toimuks populistlikku moraaliriskilist vastustundetut teadmusruumi risustamist stiilis “kaotame kõik maksud”, “rakendame teedemaksud” ning “tõstame alampalga 1000 peale” – muide täiesti teadustühised äärmiselt riskantsed säutsud – kusjuures nt Friedman’i järgi viimase postulaadi (1000) jõustamine teaduspõhiselt peaks käima nn “negatiivse tulumaksu” mehhanismi kaudu nagu ta 1962 on seda levitanud – vt ka ETA Toimeised Ühiskonnateadused 1996 autor G. Bramness) jne jne.

Jääks veel lisada et Eesti liitumisel ELiga oli risk et rahvahääletus kukub läbi (mitmete parteide magnaadid näisid politikaanselt eurovastastena (suhkur minevat kallimaks ja hiina autosid ei saa osta, sibulõunapuud tuleb maha raiuda – rääkimata Musta Luige laskumise võimalikkusest – samas makroökonoomiliselt asjatundlikult üldiselt väideti õigesti eeskätt liitumise olulisi positiivseid riske) ja läbikukkumisega väga paljud ettevalmistavad tööd/investeeringud oleks osutunud mõttetuteks. Tookord vastavaks riskianalüüsiks arvutati välja**** “kriitiline tõenäosus” – s.o võimaliku läbikukkumise maksimaalne tõenäosuse piir mille ületamisel edasised liitumisinvesteeringud oleks tulnud lõpetada (selgus et see võis olla üle 20% ning ekspertide arvates referendumi läbikukkumise võimalust hinnati oluliselt madalamaks – muide uurimus arutati läbi vastavas rahvusvahelises teaduskogus).

*) Vt nt ka The Economist 28.V 2014 arvustust “Riski business:

„Risky business“

GERD GIGERENZER, director of the Centre for Adaptive Behaviour and Cognition at the Max Planck Institute for Human Development in Berlin, is on a mission to make everyone more “risk savvy”. Mr Gigerenzer argues that being risk savvy is important if people are to protect themselves from being manipulated by politicians, doctors, financial advisers and others who claim to be looking out for their best interests.

Being risk savvy is about more than understanding the rudiments of probability theory and human psychology. It is also about being able to make judgments when the data are not available, when the situation is very complex, or when there isn’t time to calculate the risks. For these situations, there are rules of thumb, such as fear whatever your social group fears, and don’t buy a financial product that you don’t understand.

But in case anyone thinks Mr Gigerenzer is risk averse, he makes it clear in his latest book, “Risk Savvy”, that without taking risks, “innovation would end, as would fun, and courage would belong to the past”. Being risk savvy is not about avoiding risks but about understanding risks and then making an informed choice.

Mr Gigerenzer spoke to The Economist about giving up the banjo, cancer screening, and trusting your gut.

How did you first become interested in risk?

When I was a graduate student I was earning good money as a banjo-player, but I had to choose between making music my career or going into academia. I chose academia. This personal experience taught me that you can calculate risk when you play roulette in a casino, but not usually in the real world. That’s when you have to use your intuition or a rule-of-thumb, which for me was: find out what’s the most important thing for you and just go with it.

Why is it important to be risk savvy?

It is a precondition for a functioning democracy that people should not be easily frightened into surrendering their money, their welfare and their liberty. Risk savvy means being able to understand the basics of statistical reasoning. But that’s not enough. In a world of uncertainty you need rules of thumb, called heuristics. You need both those tools. And my vision is different from many of my colleagues, who think that people are basically hopeless when it comes to understanding risk, and we need to nudge them into behaviour from birth to death. My own research has shown that we can easily teach people and give them the tools. I’ve taught more than 1,000 doctors to understand health statistics, and I’ve shown that first-graders can also learn to understand Bayesian reasoning [a probability theory that uses the knowledge of prior events to predict future events]. We live in the 21st century, surrounded by complex technology, and there are things that we will not be able to anticipate. What we need is not just better technology, bigger bureaucracy and stricter laws that limit individual freedom, but risk-savvy citizens.

Your research shows that Russians are much better at understanding the risks of cancer screening than their German, French or British counterparts. Is that because the Russians are better at teaching risk in schools?

No. There is a simple explanation: they get less information and less misleading information. The problem is not that people are stupid; the problem is that the information is often presented in a misleading way. For instance, women get information about breast-cancer screening in relative risk, which looks big, rather than absolute risk. For every thousand women who don’t participate in breast cancer screening programmes, from age 50, about five will die in the next 10 years from breast cancer. And for every 1,000 women who do participate in the screening programme, it’s four. But that drop from 5 in 1,000 to 4 in 1,000 is represented as a 20% reduction in deaths. This is deliberate misinformation.

Why are we being misinformed?

There are interest groups who lobby, but also we should not underestimate the lack of knowledge about evidence in medicine. I’ve had presidents of national cancer societies in my office and found out to my surprise that they don’t know what the benefit of screening is.

You mention in your book that when a situation is complex or when the data are not available to you, you can use rules of thumb to determine risk. Can you give an example of a rule of thumb?

Assume you want to invest money and you use the Markowitz mean-variance portfolio optimisation [an investment model used by many banks]—Harry Markowitz won the Nobel Prize for that. But when Markowitz did his own investment, he didn’t use his own technique but a simple rule of thumb, which is: divide your money equally among your investments. So if you have two investment options, you invest 50-50. If you have three options, it’s 33.3%. This rule of thumb is called 1/N. You don’t have to do all these calculations and estimations. A number of studies show that investing this way typically does better than the Nobel prize-winning method.

You also talk about an unconscious rules of thumb, or intuition. What role does intuition play in assessing risk?

In our society, intuition is looked upon as suspicious. I have done a number of studies with large international companies and asked the decision-maker how often do you make important decisions with your gut? A gut decision is not arbitrary, or a sixth sense. It’s based on lots of experience, but it is in the unconscious. On average, we found that about 50% of all these big decisions were gut decisions. But the same managers would not admit this in public. There is anxiety because they could be made responsible. And intuition, even if it’s better than calculation, has a bad name in our society.

Aren’t anecdotes about people who successfully used their intuition to make an important decision prone to survivorship bias? We only get to hear their success stories because they survived or their company survived or were successful in some way.

My point is not that intuition is always superior. My point is that we need more tools. And for me intuition is equal with statistical calculations. The real question is: can we identify the problem where it is better to go after your first gut feeling, and the type of problem where it’s better to go and think about the problem and collect data? So a strategy like a heuristic is not better or worse, it’s just that you need to figure out where it works.

Do you think that people who smoke or overeat do so because they don’t understand the risks involved?

Smoking is a special case because you get addicted. What can you do? In my opinion, the one thing that is not followed through and tried is to work with children in first grade. There’s no point in telling a 15-year-old to stop smoking, it is too late. You have to start early with some kind of health literacy. And also teach them the psychological mechanisms that the tobacco industry uses to make them want to smoke, like advertisements. And if we can only reach 10% or 20% of these children we can save more people from cancer than if you put the same amount of money in the development of the next cancer drug. But almost no money is given for prevention.

I’m sure people who overeat are well aware of the risks of being overweight or obese.

In the good old times when I was growing up there were rules of thumb like: eat three meals a day and nothing in between. So that helps. But you need to start educating early, before puberty when the habits are shaped, and introduce a number of rules that people understand and can use themselves to make them strong, rather than repairing them at the other end.

How different would the world be if we were all a bit more risk savvy?

It would be much more like a real democracy where people are informed and make their own choices. We would have a democracy where people would learn to ask questions and learn to be critical, and not be willing to have their money taken away, their health endangered, and their liberty taken away.

**) Vt nt: http://ec.europa.eu/taxation_customs/resources/documents/taxation/company_tax/transparency/com_2015_136_en.pdf

***)

Ennuste,Ü. 2014. Towards Special Methodological Problems of Macro-Optimal Sociocybernetic International Economic Sanctioning Coordination Modelling: Introductory Remarks oPreliminary Postulates and Conjectures. – Baltic Journal of European Studies Tallinn University of Technology (ISSN 2228-0588), Vol. 4, No. 2 (17), 150-158:

http://www.degruyter.com/view/j/bjes.2014.4.issue-2/bjes-2014-0021/bjes-2014-0021.xml?format=INT

****) Ennuste, Ü. and Rajasalu, T. 2002. Critical Probability of the EU Eastern Enlargement Project’s Institutional Failure: Aspects of Calibrated Economic Impacts of the Failure. In: Aksel Kirch and Juhan Sillaste (eds.) Monitoring Preparations of Transition Countries for EU-Accession. 4th International Conference 4-6 October, 2002 Pärnu, Estonia, The Institute for European Studies, Tallinn, 212-227: http://www.ies.ee/iesp/ennuste.pdf

märts 25, 2015 - Posted by | Uncategorized

Kommentaare veel pole.

Lisa kommentaar

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Muuda )

Twitter picture

You are commenting using your Twitter account. Log Out / Muuda )

Facebook photo

You are commenting using your Facebook account. Log Out / Muuda )

Google+ photo

You are commenting using your Google+ account. Log Out / Muuda )

Connecting to %s

%d bloggers like this: