Ilija Barukcic

C a u s a l i t y.

A theory of energy, time and space.

ISBN: 978-1-4092-2952-0

Volume 1. Fifth Edition 2008.

552 pages. Paperback. $69.50

ISBN: 978-1-4092-2952-0

Volume 2. Fifth Edition 2008.

320 pages. Paperback. $49.50

Publishing company: Lulu : http://www.lulu.de/

... back to Welcome

Last Update:

June 09th, 2007.

Recommend this page to a friend

 

 

Please, choose among ...

Conferences

Marketing : Web Stores, Fax ...

(Book) Review

Downloads

Comments On The Book Reviews

Discussion With The Reader

Causation: Background

Philosopher/Philosophy

Mathematics, Statistics, Probability

Causality: Workshop, Conference, Congress ...

Job Openings

Causality: E Mail List

Contact

 

 

 

 

Conferences

2006

Biometrics

XXIII International Biometric Conference, Montreal, Canada.

http://www.ibc2006.org/e/01-home_e.shtml

This conference brings together statisticians and others interested in the development and application of statistical and mathematical methods for the biological sciences.

My contribution:

New Method for Calculating Causal Relationships.

I. Barukcic, Jever, Germany

The Scientific Program (pdf) is now available ! ->->->-> page 41.

http://www.ibc2006.org/pdf/sciprog-1-May-2006.pdf.

http://www.ibc2006.org/pdf/sciprog-29-June-2006.pdf, p. 49, p. 53.

http://www.ibc2006.org/pdf/sciprog-29-June-2006.pdf.

2007

Quantum Mechanics

Quantum Theory:

Reconsideration of Foundations - 4

June 11-16, 2007

Växjö University,

Sweden.

http://www.vxu.se/msi/icmm/qtrf4/program.pdf p. 6

My contribution:

Bell's Theorem - A fallacy of the excluded middle.

http://www.vxu.se/msi/icmm/qtrf4/program.pdf p. 6

Back to top

 

Downloads

Second Edition 2006.

Contents:

Ilija Barukcic. Causality. New Statistical Methods.

Second Edition, 2006. (*.pdf file)

September, 7th 2006:

The second Edition is available!

The second Editions of my book is available. If you should have any problems to get my book, please use the fax order (download it from this web page) or please feel free to contact me or the publishing company. It is possible to help you.

Last Update: 07.09.2006.

First Edition 2005.

I am sorry, the first Edition 2005 is no longer available.

Erratum first Edition:

Ilija Barukcic Causality. New Statistical Methods. 2005. (*.pdf file).

 

Pictures: Ilija Barukcic. Die Kausalität. 1997. (*.pdf file)

Ilija Barukcic. (*.pdf file)

Get Adobe Reader at

http://www.adobe.de/products/acrobat/readstep2_allversions.html

to study the *.pdf files above.

Last revision: February 22th 2006.

Back to top

 

 

 

(Book) Reviews

2006

Alain Vandal

XXIII International Biometric Conference

Montreal Canada July 16-21 2006

Presentation of the new methods at the conference in Montreal.

2006

Marry E Thompson

Book Review Marry E Thompson

ISI - INTERNATIONAL STATISTICAL INSTITUTE

http://isi.cbs.nl/

ISI - Short Book Reviews

http://isi.cbs.nl/sbr/sbrRev2006.htm

...

Back to top

 

 

 

Comments On The

(Book) Reviews

2006

Comments on Thompson's Book Review, April 2006

Get a book for free for a Book Review

(at http://www.bod.de/presse/rezensionsexemplare.html ):

Condition: A Book Review must be published!

Help me on request for Book Review (*.GIF File)

Mail this to a friend

Back to top

 

Discussions

( With The Reader )

Mail a question to Barukcic

26.05.2006

Jack Himelright:

Karl Popper - The principle of falsifiability

29.10.2006:

Lofti A Zadeh

Causation, Data and Post hoc, ergo propter hoc

Back to top

 

Marketing

Mail: Order Barukcic's Causality.

Please contact me if you need some calculations on your data.

 

Webstores

Libri

Amazon

Bol

Spiegel

Buch

FAZ

Hugendubel

Buch24

Thalia

Order the Book ( Fax )

Back to top

 

 

   

Causation: Background

The central idea behind my publication and the aim to characterise the relationship between cause and effect using the tools of classical logic, probability theory, philosophy, statistics is based on the fact, that

causes have influence on the probabilities of their effects

(Suppes, Hesslow, ...).

But my publication is not simply a new probabilistic theory of causation, it is much more then this. The basic work on the unified field theory is done too. Einstein's and Heisenberg's equations can be derived directly from my equations. Einstein's field equations have to do a lot with causation. Causality is absolutely compatible with Einstein's Relativity and with Heisenberg's Uncertainty Principle.

This is proofed by me very precisely and is published in the second Edition of my monograph above.

Causality does not only provide new methods for extracting causal relationships from experimental and nonexperimental data.

Causality is an outstanding contribution

to the Philosophy of Science,

to Classical/Mathematical Logic and Scientific Methods,

to Mathematical Statistics and Probability and

to Physics and

to ...

Finally,

a unified mathematical and statistical model of causation

is mathematically proofed and available.

In this sense, the asymmetry of causation is often understood that way,

that causes must precede their effects in time. Only, can an effect as such occur in the absence of a cause? Can a random variable be a cause of an effect in the absence of an effect? To what extent may a cause as such be independent from an effect? Until my publication, it was generally unclear and mathematically not proofed, can or to what extent can causes precede their effects in time?

Probabilistic theories of causation do indeed require that a cause has influence on the probability of its effect. Are such theories of causation compatible with indeterminism at all or is causation at the end incompatible with probability and at the end with indeterminism?

Is it necessary to develop a theory of causation that does not presuppose determinism? My book gives substantial mathematical and philosophical answers to the mathematical and philosophical problem of causality.

 

 

 

 

 

 

The

mathematical

and

statistical

problem

of

causality

is

definitely

s o l v e d !

Causality is mathematised,

the probabilistic approach to causation has won the race.

Based on the deterministic relationship between cause and effect,

Boole's, mathematical and classical logic, at the top of the formal development of

the ( statistical and probabilistic ) theory of causation,

the mathematical formula of

the causal relationship c

is derived and mathematically proofed!

Making causal inferences on the basis of experimental and nonexperimental data is possible, it doesn't matter if the random variables have a discrete, a continuous or

a mixed distribution. It doesn't matter if the mathematical formula of the causal relationship c is used in Epidemiology, in Sociology, in Biostatistics, in Economy, in Medicine, in Nuclear physics, in ... etc.

Methods for the calculation of the critical value of the causal relationship c are developed. Methods for the calculation of the p value, of beta and power of the causal relationship c (left, right, two sided) are developed. Thus, a so called test statistic is available to proof data for causality.

In a reanalysis based on the published data of Naoumi Uemura et al, N Engl J Med 345

(2001) 11, 784, i have proofed, that the bacterium

Helicobacter pylori is the cause of human gastric cancer

( p value = 0.001999831220, power > 0.85 ).

This was published by me August 2005 and thus long before the Nobel Prize announcement by Professor Hans Jörnvall, Secretary of the Nobel Committee for Physiology or Medicine, on October 3, 2005, that the Nobel Prize in Physiology or Medicine 2005 will be awarded to Barry J. Marshall and J. Robin Warren

"for their discovery of the bacterium Helicobacter pylori and its role in gastritis and peptic ulcer disease"

( -> http://nobelprize.org/medicine/laureates/2005/index.html ).

In so far,

it is me who has found/identified

the cause of

human gastric cancer,

based on the data of Naoumi Uemura et al,

and no other then me!

Comments On The Study Of Wong et al.

 

 In a reanalysis based on the data of Mathilde Bonnet et al. ( Journal of the National Cancer

Institute, Vol. 91, No 16 (1999) 1376-1381), i have proofed, that

Epstein-Barr virus (EBV) is a cause and not the cause of human invasive breast carcinoma ( p value = 0.000321, power > 0.99 ).

According to this investigation, not all, but just about 50 % of all human invasive breast carcinoma are definitely caused by Epstein-Barr virus.

The one and only assumption based on the determination of cause and effect is that the random variables must be Bernoulli distributed (or Bernoulli distributed in character). Thus, any density function, any (empirical) cumulative distribution function, any probability mass function etc. of random variables can be used for calculating causal relationships without any restriction.

Needless to say, the quality of data must be assured.

My new methods must be distinguished from other publications especially from the procedures of Judea Pearl, Peter Spirtes, Clark Glymour and Richard Scheines. To many explicit assumptions that are rarely satisfied in realistic cases must be made before the procedures of Judea Pearl, Peter Spirtes, Clark Glymour and Richard Scheines can yield results, if at all. In so far, at the end, this procedures are somehow not that much useful for solving causality problems.

Back to top

 

 

Causation vs. Correlation

In contrast to causation, the central idea behind correlation is

the equivalence of a Quantity X and a Quantity Y

or in other words

Quantity X = Quantity Y

Or

Y = X.

Let E(X) denote the expectation values of X.

Let E(Y) denote the expectation values of Y.

Thus, according to Kolmogoroff, it is true that

E( Y ) = E ( X ).

After some calculations, we do obtain the formula of the correlation coefficient of

a population as (Barukcic, Causality. ..., p. 340-343)

r = s ( Y, X ) / ( s ( Y ) * s ( X ) ).

The correlation coefficient states only, that there is a linear relationship between

a Quantity X and another Quantity Y and not that a Quantity X is the cause

of a Quantity Y.

It is known, that correlation analysis is based on the cum hoc, ergo propter hoc

fallacy.

Under certain circumstances, a Quantity X can be a cause of a Quantity Y,

only this can not be proofed with the correlation coefficient.

In this case, it is necessary to use my methods to succeed.

Correlation has indeed nothing to do with causation or with the new mathematical formula of the causal relationship c.

Back to top

 

Conditions vs. Counterfactuals

Conditio per quam vs. risk ratio

In Epidemiology, Biostatistics etc. the risk ratio is a very important relationship.

Thus let us regard the following contingency table.

The risk ratio (RR) is defined as

RR = ( a / ( a + b )) / ( c / ( c + d ) ).

Recall:

Exposure proportion in cases = a / ( a + c ).

Exposure proportion in controls = b / ( b + d ).

Odds ratio : OR = ( ( a * d ) / ( b * c ) ).

Let us perform a thought experiment

(from the German term Gedankenexperiment, coined by Ernst Mach)

Please allow me to presuppose, that the most of the reader will be able to remember

that it is raining sometimes. On the other hand, the most of us will have observed,

that the rain changes somehow other random variables too. Thus, let us ask our self:

is there a relationship between raining and a street that is wet?

Or in other words,

is the rain a risk factor of a wet street?

In this case the following should hold true:

Let us perform n=100 Bernoulli trials somewhere out there under haven where it is raining very often (f. e. the equator) and let us obtain f. e. the following result.

The risk ratio RR in this case is

RR = ( a / ( a + b )) / ( c / ( c + d ) ) or

RR = ( 50/50)/ (49/50) = 1,02040816326530612244897959183673.

The risk ratio RR = 1,02040816326530612244897959183673

or in other words,

there is no effect of the rain on the street,

there is no significant association between raining and a street that is wet.

The street was exposed to the rain, but besides of this

the rain as such had not any effect on the street,

according to the investigation above and the risk ratio relationship.

Do we really have another choice if we rely on the data above and on the risk ration then to conclude that

the rain is not a risk factor of a wet street!

But please stop for a while. Common sense must tell us, that this result is of course far away from reality.

It is generally known, that when the rain falls, then the street is wet. The rain in this sense must be a risk factor of a wet street, absolutely. Thus, it is easily seen, that the capabilities of the risk ratio are very limited. We need a more reliable relationships, some that does not depend that much upon study design.

Thus, let us calculate the probability of the conditio per quam relationship

of the contingency table above. We obtain

p ( It is Raining -> The street is wet ) = 1,

an absolutely significant result.

Thus, we can say, it is true, that when it is raining, then the street is wet.

The same data allow us two conclusions that are contrary!

Let us perform another n=100 Bernoulli trials somewhere out there ( e. g. desert like Sahara without that much rain) under haven and let us obtain f. e. the following result.

The risk ratio RR in this case is

RR = ( a / ( a + b )) / ( c / ( c + d ) ) or

RR = ( 1/1)/ (98/99) = 1,01020408163265306122448979591837.

The risk ratio RR = 1,01020408163265306122448979591837

and once again, there is no effect of the rain on the street,

there is no significant association

A second investigation proofed the relationship between

raining and a wet street and the result is the same

there is no significant association. In so far, it seems to me

that it is time to accept, that

the rain is not a risk factor of a wet street!

What is wrong?

This two simple examples show us the dramatic errors that can happen, when we use some old relationships, especially the risk ratio. The risk ratio as such and some other similar old relationships too

depend to much upon study design or

on how cases and controls are sampled from

the source population.

They can lead to totally wrong results!

Let us calculate the probability of the conditio per quam relationship

of the contingency table above, thus we get

p ( It is Raining -> The street is wet ) = 1,

an absolutely significant result.

The conditio per quam is known since thousands of years and is not depending that much upon study design. The conditio per quam relationship is much more reliable then the risk ratio! This relationship is developed by me, mathematically proofed and can be used!

Back to top

 

Conditio sine qua non: Without A no B

Sir William Richard Shaboe Doll, ( Doll R, Hill AB. Smoking and carcinoma of the lung. Preliminary report, British Medical Journal, 2: (1950) 739-748 ), a British epidemiologist, was the first in the world who investigated in a historically important case-control study ( retrospektiv, Fall Kontroll Studie) the relationship between smoking and lung cancer. He found that 647 of 649 lung cancer cases were smokers. In contrast that, 622 of 649 non-cancer controls were smokers. Let us have a look on these data in a 2-by-2 table.

Is there a deterministic relationship between

smoking and lung cancer?

 

The odds ratio OR can be calculated as

OR = ( ( a * d ) / ( b * c ) )

OR = ( (647 * 27 ) / (622 * 2 ) )

OR = ( (17469 ) / (1244 ) )

OR = 14,04..

This result indicates a strong association.

Let us reanalyse the data above and calculate

the conditio sine qua non relationship

( Without <condition> no <conditioned>).

 

 

Hypothesis:

Ho: p( Smoking <- Lung Cancer ) = 1.0

HA: p( Smoking <- Lung Cancer ) < 1.0

Alpha: 0.05, one sided test.

Material and Methods

Results

p( Smoking <- Lung Cancer ) = ( ... ) / ...

p( Smoking <- Lung Cancer ) = 0,99845916795069337442218798151002

p( Smoking <- Lung Cancer ) critical = ...

p( Smoking <- Lung Cancer ) critical = ...

p( Smoking <- Lung Cancer ) critical = 0,988586208889688259139412043748017

Discussion

The data above do support the Null hypothesis, that

WITHOUT SMOKING NO LUNG CANCER

because

p( Smoking <- Lung Cancer ) > p( Smoking <- Lung Cancer ) critical.

 

Let us use the data above just for demonstration purposes

to calculate the causal relationship.

Hypothesis:

Ho: c ( Smoking => Lung Cancer ) < 0

HA: c ( Smoking => Lung Cancer ) > 0

Alpha: 0.05, one sided test.

Results

c ( Smoking => Lung Cancer ) = + 0,130319738149084270086198603727516.

-> Z Value = 4,69512917056785193772410008292137 -> p Value < 0,000005

c ( Smoking => Lung Cancer ) critical = ... / ...

c ( Smoking => Lung Cancer ) critical = ... / ...

c ( Smoking => Lung Cancer ) critical = 0,0456551644412469634423518250079336

Z Beta = ... - ...

Z Beta = -3,05027554056785193772410008292137

p Beta ~ 0.001

Power = 1 - p Beta = 1 - 0.001 = 0,999

 

 

Since

c ( Smoking => Lung Cancer ) > c ( Smoking => Lung Cancer ) critical

we must reject the Null hypothesis and accept the alternative hypothesis.

There is a highly significant causal relationship between

smoking and lung cancer ( p Value < 0,000005, power > 0.99 ).

Discussion

Thus, using the new mathematically proofed methods we can state.

Firstly.

Smoking is a conditio sine qua non of lung cancer or that is to say

without smoking

no lung cancer.

 

Please, pay attention. Smoking is not a conditio per quam of a lung cancer

or in other words it is not that way, that when you smoke, then you must

become lung cancer.

Without smoking, a patient can't develop lung cancer.

Secondly.

Smoking is not only a condition of lung cancer, it is much more then this.

Smoking is at the same time the cause and not one cause of lung cancer too

( p Value < 0,000005, power > 0.99 ), a highly significant result.

Recall, the statistical power of a test is the probability of correctly rejecting

a false H0 under certain distributional assumptions.

Back to top

 

 

 

 

Counterfactuals

Counterfactual Theories of Causation

are based more or less on the following assumption.

Let c (=smoking) be the cause of the effect e (= lung cancer).

According to the counterfactual theories of causation (David Hume, Lewis),

if the cause c had not occurred,

then the effect e would not have occurred too.

In so far, this is not absolutely wrong. Still, there are many doubts on the adequacy of the analysis of causation in terms of counterfactuals. According to the counterfactual theories of causation, one cause of an effect is any event for which an effect would not have occurred.

Example.

Let a person develop lung cancer because of smoking. In terms of

counterfactuals it is indeed true that if this person had not smoked

this person would not have got lung cancer. In so far, not smoking is indeed thus an event for which an effect would not have occurred (not lung cancer).

Only, what would happen, if the person above had not been born at all. It appears reasonable to accept, that if the person above had not been born or if the person above had not possessed lungs, the person above would not have got lung cancer too. Consequently, according to the counterfactual theories of causation, if the person above had not been born is equally an event for which an effect (lung cancer) would not have occurred and stands thus according to the counterfactual theories of causation in causal relation to lung cancer.

Only, how can you get lung cancer, if you not born?

On the other hand, there are a lot of persons who are born and they do not develop lung cancer. Being born as such is not a cause of lung cancer. We assumed above that smoking is the cause of lung cancer. According to the counterfactual theories of causation it is not smoking that is the cause of lung cancer, it is the fact of being born. According to the counterfactual theories of causation smoking is the cause of lung cancer and at the same time smoking is not a cause of lung cancer (being born is the cause) and both is true. We arrived at an impossible result.

This is a logical contradiction.

This simple reductio ad absurdum argues any relevance of the counterfactual approach to causation.

Counterfactual Theories of Causation

are generating to many absurd results and

useless for solving

the problem of causation.

Back to top

 

 

Boolean Algebra vs. Fuzzy Logic

The fuzzy logic as such is not compatible with the fundamental laws of probability

theory. This is proofed by me very precisely.

Methods to use the rules of Boolean algebra for the values between 0 and 1 according to the fundamental laws of probability are developed and mathematically proofed.

A Method to use the Quine-McCluskey Method on discrete and continuous

random variables is developed and mathematically proofed.

Thus, redundant variables can easily be identified, with the possibility to reduced complicated multirelationships to a few and simple one!

Back to top

 

  

Time series

Many new Time series relationships are developed

f. e. After a comes b ( After the day comes the night ...),

a must before b,

a before b.

...

Back to top

 

 

Structural equation modelling (SEM)

Structural equation modelling (SEM) has dominated causal analysis for a long time. But even SEM researchers (Muthen 1987) have difficulties of finding the causal content of SEM. In so far, structural equation modelling (SEM) is widely (mis)used in biology,

sociology, econometrics and other sciences. At the end, this rapidly growing and extremely complicated statistical machinery focuses today more or less only on model fitting.

But even here, SEM fails. The widespread (mis) use of this complex and often poorly understood analytic methods is characterised by the trial to represent a causal theory by a path diagram and translation of this into a structural equation model like

y = (ß*x) + e.

There are so many concerns associated with structural equation modelling,

where to start?

Firstly.

The world as such cannot be pressed into a linear or structural equation

Model or being considered as only deterministic.

Secondly.

Structural equation modelling (SEM) is based on an absolutely insufficient

approach to causation.

Thirdly.

Structural equation modelling (SEM) is incompatible with quantum mechanics

and belongs today already to the past.

Expensive training seminars are more or less necessary to learn this statistical techniques which at the end have nothing to do with causation. It appears to me that only a few individuals have benefited from using these statistical techniques (courses in structural equation modelling are sometimes very expensive ).

Conclusion. Structural equation modelling (SEM) is a complicated statistical machinery

and useless for discovering cause-effect relationships between variables of interest.

The SEM starting point is not correct.

-> Cartwright N., (1995). Probabilities and Experiments, Journal of Econometrics 67: 47-59.

Structural equation modelling (SEM) in this sense is not worth loosing time, money

and effort!

...

Back to top

 

 

Philosopher/Philosophy

Georg Wilhelm Friedrich Hegel

International Virtual Library

http://www.philo.de/VirtualLibrary/14.de.htm

http://plato.stanford.edu/entries/hegel/

The International Hegel-Society (Internationale Hegel-Gesellschaft e.V.)

Home: http://www.hegel-gesellschaft.de/

The Hegel Society of Great Britain (HSGB)

Home: http://www.hsgb.group.shef.ac.uk/index.html

Links: http://www.hsgb.group.shef.ac.uk/links.html

The Hegel Society of America

http://www.hegel.org/index.html

Science of Logic

http://www.marxists.org/reference/archive/hegel/index.htm

David Hume -> http://plato.stanford.edu/entries/hume/

David Hume, An Enquiry Concerning Human Understanding (1748).

Immanuel Kant: Akademieausgabe von Immanuel Kants Gesammelten Werken

-> http://www.ikp.uni-bonn.de/kant/verzeichnisse-gesamt.html 

Suppes, Patrick. -> http://www-philosophy.stanford.edu/fss/ps.html

1970. A Probabilistic Theory of Causality. Amsterdam: North-Holland.

Germund Hesslow -> http://www.mphy.lu.se/avd/nf/hesslow/philosophy/TWONOTES.htm

Homepage -> http://www.mphy.lu.se/avd/nf/hesslow/

Philosophy - selected publications -> http://www.mphy.lu.se/avd/nf/hesslow/

Wolfgang Spohn

-> http://www.uni-konstanz.de/FuF/Philo/Philosophie/Spohn/spohn.shtml

Papers -> http://www.uni-konstanz.de/FuF/Philo/Philosophie/Spohn/spohn_papers.shtml

David Kellogg Lewis (*September 28, 1941 + October 14, 2001)

-> http://en.wikipedia.org/wiki/David_Lewis_(philosopher)

Counterfactuals (1973 [revised printing 1986]; Blackwell & Harvard U.P.)

Ellery Eells (1953 - + August 10, 2006 ). -> http://philosophy.wisc.edu/eells/

Ellery Eells, Probabilistic Causality. Cambridge University Press, 1991.

Jon Williamson -> http://www.kent.ac.uk/secl/philosophy/Staff/williamson.htm

University of Kent, Canterbury.

Nancy Cartwright -> http://personal.lse.ac.uk/cartwrig/Default.htm

Nancy Cartwright Recent Papers -> http://personal.lse.ac.uk/cartwrig/Papers.htm

University of California San Diego.

Hitchcock, Christopher -> http://www.hss.caltech.edu/people/faculty/cricky

Hitchcock, Christopher. 1993. "A Generalized Probabilistic Theory of Causal Relevance." Synthese 97: 335-64.

John David Collins -> http://collins.philo.columbia.edu/index.html

Ned Hall -> MIT

L. A. Paul -> http://www.u.arizona.edu/~lapaul/

John Collins, Ned Hall and L. A. Paul
Causation and Counterfactuals: MIT Press, June 2004

ISBN 0-262-53256-5. MIT PRESS.
Iain Martel -> http://www.erin.utoronto.ca/~imartel/Iain/index.htm

Douglas Kutach -> http://www.brown.edu/Departments/Philosophy/Douglas_Kutach/index.html

Robin Cowan -> http://www.cgl.uwaterloo.ca/~racowan/cause.html
Mario J. Rizzo -> http://www.cgl.uwaterloo.ca/~racowan/cause.html

Paul Noordhof -> http://www.nottingham.ac.uk/philosophy/staff/paul-noordhof.htm

Dialectical Materialism

S Barker

Counterfactuals, probabilistic counterfactuals and causation

Mind 1999 108(431):427-469

Charles A. Mercier:

Causation 1916: -> http://www.geocities.com/freasoner_2000/cause.htm

Causation 1916 -> http://evans-experientialism.freewebspace.com/mercier02.htm

Logic MCMXII -> http://www.geocities.com/freasoner_2000/logic.htm

Logic MCMXII *.pdf: -> http://www.cwru.edu/edocs/8/244.pdf

Back to top

 

 

The Causal Markov Condition

N. Cartwright: "Against Modualrity, The Causal Markov Condition and Link Between the Two: Comments on Housman and Woodward"

http://personal.lse.ac.uk/cartwrig/Papers.htm

I. Martel: Indeterminism and the Causal Markov Condition

http://www.erin.utoronto.ca/~imartel/Iain/Research.htm

D. Steel: Indeterminism and the Causal Markov Condition

http://www.msu.edu/user/steel/Ind&CMC.pdf

Overview:

http://classes.engr.oregonstate.edu/eecs/winter2006/cs539/slides/causalnetworks.pdf

Berkovitz:

On Causal Inference in Determinism and indeterminism

http://www.umbc.edu/philosophy/berkovitz/papers/determinism.pdf

Back to top

 

 The Principle of the Common Cause

Berkovitz: On Causal Inference in Determinism and indeterminism

http://www.umbc.edu/philosophy/berkovitz/papers/determinism.pdf

I. Martel: Indeterminism and the Causal Markov Condition

http://www.uni-konstanz.de/ppm/workshop2/Martel_final.pdf

Wolfgang Spohn: On Reichenbach's Principle of the Common Cause

http://www.uni-konstanz.de/FuF/Philo/Philosophie/Mitarbeiter/spohn_files/ wspohn22.pdf

Back to top

Philosophen und Philosophinnen

http://philo.at/philpages/philosophen.html ( Prof. Herbert Hrachovec, Wien )

PhilSearch -> http://search.freefind.com/

Hegel-Links -> http://hegel.net/en/links.htm

Back to top

Centre for Philosophy of Natural and Social Science :

http://www.lse.ac.uk/collections/CPNSS/Default.htm

London School of Economics

http://www.lse.ac.uk/

3rd International Summer School 2004. University of Konstanz

http://www.uni-konstanz.de/ppm/summerschool2004/

Back to top

 

 

Mathematics, Statistics, Probability

James Robins

http://www.hsph.harvard.edu/facres/rbns.html

Research

http://www.biostat.harvard.edu/~robins/research.html

Judea Pearl

http://bayes.cs.ucla.edu/jp_home.html

Judea Pearl: Causality: models, reasoning and inference (Cambridge University Press, 2000). ISBN: 0521773628

Peter Spirtes

http://www.hss.cmu.edu/philosophy/faculty-spirtes.php

Clark Glymour

http://www.informatics.indiana.edu/ra/presenters/glymour.asp

Richard Scheines

http://www.hss.cmu.edu/philosophy/faculty-scheines.php

Causation, Prediction, and Search, Mit Pr . ISBN: 0262194406

The authors use the formalism of the Bayes networks to turn observations into causal knowledge but without the final success.

Causality Lab:

http://www.phil.cmu.edu/projects/causality-lab/

Richard E. Neapolitan

http://www.neiu.edu/~reneapol/renpag1.htm

Computer Science Dept., Northeastern Illinois University

Larry Wasserman

http://lib.stat.cmu.edu/~larry/

Carnegie Mellon University

Mark van der Laan

http://sph.berkeley.edu/faculty/vanderlaan.htm

M. Elizabeth Halloran

http://www.sph.emory.edu/~mehallo/ Emory University

Thomas Richardson

http://www.stat.washington.edu/tsr/ University of Washington

Nanny Wermuth

http://www.cs.chalmers.se/~wermuth/ Chalmers University of Gothenburg

Rolf Haeni

Bern University

Phillip David

University College London, DeGroot Prize, ... Bayesian Networks

JEFFREY M. ALBERT

http://epbiwww.case.edu/faculty/albert.html

Dominik Janzing

http://www.ira.uka.de/I3V_HTML/MITARBEITER/2335

http://iaks-www.ira.uka.de/ Karlsruhe

Back to top

 

 

Mathematics, Statistics, Probability: Org

International Statistical Institute (ISI)

Daniel Berze, Director

http://isi.cbs.nl/perm.htm

http://isi.cbs.nl/

International Biometric Society (IBS)

http://www.tibs.org/

Bernoulli Society For Mathematical Statistics And Probability

http://isi.cbs.nl/BS/bshome.htm

American Mathematical Society

http://www.ams.org/

European Mathematical Society

http://www.emis.de/

International Epidemiology Association

http://www.dundee.ac.uk/iea/

American College of Epidemiology

http://www.acepidemiology2.org/societies/AAP-ES.asp

International Society for Clinical Biostatistics

http://www.iscb.org/

Institute of Mathematical Statistics

http://www.imstat.org/

American Statistical Association

http://www.amstat.org/

Deutsche Statistische Gesellschaft (DStatG)

http://www.dstatg.de/

Deutsche Gesellschaft für Medizinische Informatik,Biometrie und Epidemiologie

http://www.gmds.de/

Royal Statistical Society (RSS)

http://www.rss.org.uk/

Biomathematics and Statistics Scotland

http://www.bioss.sari.ac.uk/

Statistical Society of Canada

http://www.ssc.ca/

Statistical Society of Australia

http://www.statsoc.org.au/

The Probability Web

http://www.mathcs.carleton.edu/probweb/probweb.html

Statistical/Biostatistical Organisations

http://biostat-resources.com/id4.htm

Back to top

Causality: Conference, Congress

Second International Congress For The Unity Of Science, 1936 Copenhagen

http://www.tu-harburg.de/rzt/rzt/it/QM/schilpp.html

Multidisciplinary Conference on Causality and Statistics, August 2000

http://isi.cbs.nl/Bnews/00a/bn_5.html

Joint International Conference on Cognitive Science

Sydney, Australia. 13 - 17 July, 2003

http://hps.arts.unsw.edu.au/cogsci_conf/index.html

Workshop on Causality and Causal Discovery May 16th 2004

http://cs.uwindsor.ca/~ai04/causality.htm

Vinterkonference 2006

http://www.matstat.umu.se/aktuellt/vinterkonf/v-konf-06/in_english.asp

CAUSALITY AND PROBABILITY IN THE SCIENCES

14-16 June 2006, Keynes College Lecture Room 5,

University of Kent, Canterbury, UK

http://www.kent.ac.uk/secl/philosophy/jw/2006/capits.htm

Empirical Evaluation of Causality, 26 June - 30 June, 2006

http://www.isr.umich.edu/cps/eitm/eitm2006/causality2006.html

Conferences

with my participation.

XXIII International Biometric Conference

(IBC 2006) July 16 - 21 2006,

Montreal, Canada.

http://www.ibc2006.org/pdf/sciprog-29-June-2006.pdf

http://www.ibc2006.org/pdf/sciprog-29-June-2006.pdf, p. 49, p. 53,

Quantum mechanics: Reconsideration of Foundations - 4

June 11 - 16, 2007

Växjö University - Sweden.

Back to top

 

 

Relativistic Causality

Conference in Honour of Abner Shimony

http://perimeterinstitute.ca/activities/scientific/shimony/index.php

http://en.wikipedia.org/wiki/Abner_Shimony

Perimeter Institute, Waterloo, Ontario, Canada

July 18 - 21, 2006

CHSH inequality

-> http://en.wikipedia.org/wiki/CHSH_Bell_test

Bell's inequality

http://en.wikipedia.org/wiki/Bell_inequality

Barukcic has refuted Bell's Theorem.

Local hidden local variables exist.

Mail this to a friend

Back to top

 

 

Job Openings

International Statistical Institute (ISI):

http://isi.cbs.nl/AD.htm

International Biometric Society

http://www.tibs.org/advertising.htm

Germany

Internationale Biometrische Gesellschaft

Deutsche Region

http://www.biometrische-gesellschaft.de/

Deutsche Gesellschaft für Medizinische Informatik,Biometrie und Epidemiologie

http://www.gmds.de/stellenboerse/stellenboerse.php

Weitere Stellenbörsen

http://www.gmds.de/stellenboerse/weitere_links.php

http://www.bfarm.de/de/DasBfArM/jobs/index.php

http://www.dimdi.de/

http://www.gsf.de/epi/de/index_stellen.html

Back to top

 

Causality: E Mail List

I am managing an email list about causality.

If you are not already on this list

please drop some lines at

Barukcic at t-online dot de

You can be removed from this list the way you like it.

There are no obligations or costs for you.

I'm sorry, you can't know who is already on this email list.

Back to top

 

 

 

Contact

E-Mail:

Barukcic < a_t > t-online < dot > de

Handy:

( 00 49 ) - ( 0 ) 1 51 - 100 232 58

Phone:

( 00 49 ) - 44 23 - 99 11 11

GMT:

+ 1 h.

Germany.

Back to top

 

 

 

AWARDS FOR RESEARCHERS

Nobel prize

http://nobelprize.org/

Balzan Prize

http://www.balzan.com/

International Balzan Foundation

CH-8022 Zürich

mailto: info@balzan.com

Lakatos Award

Philosophy of Science

The London School of Economics and Political Science

http://www.lse.ac.uk/

http://www.lse.ac.uk/collections/

The International Mathematical Union

http://www.mathunion.org/

The Fields Medal

http://www.fields.utoronto.ca/

Rolf Nevanlinna Prize

http://www.mathunion.org/medals/Nevanlinna/index.html

Carl Friedrich Gauss Prize for Applications of Mathematics

http://www.mathunion.org/General/Prizes/Gauss/index.html

The European Science Foundation

The DeGroot Prize

http://www.bayesian.org/awards/DeGrootPrize.html

The Prize consists of an award of $1500 and a commemorative plaque.

The Prize is awarded every two years.

All books published no earlier than 5 years prior to the year of the DeGroot Prize competition may be considered.

Authors may self-nominate, or may be nominated by another individual or individuals.

MORTIMER SPIEGELMAN AWARD

http://www.aphastat.dpcp.org/Spiegelman%20winners.htm

Marvin Zelen Leadership Award in Statistical Science

http://www.biostat.harvard.edu/events/awards/zelen/index.html#nominations

Awards presented at the ASA Presidential Address Session

Wilks Memorial Award

http://www.amstat.org/awards/index.cfm?fuseaction=wilks

The deadline for nominations is April 1

Gottfried E. Noether Awards

http://www.amstat.org/awards/index.cfm?fuseaction=noether (younger than 35 years of age)

Outstanding Statistical Application

http://www.amstat.org/awards/index.cfm?fuseaction=outstanding-application

The Award consists of a certificate and $1,000 cash prize.

Nominations should be received by April 1, 2006.

Back to top

 

 

 

Ein Hinweis zu den gelinkten Seiten.

Für den Inhalt der gelinkten Seiten Dritter übernehme ich grundsätzlich keine Haftung und auch keine Gewähr. Der Inhalt der gelinkten Seiten Dritter kann sich auch ändern ohne das ich das ständig überprüfe und an dieser Web-page zeitnah anpassen kann.

Hinsichtlich des Inhalts der gelinkten Seiten Dritter ist selbstverständlich davon

auszugehen, das ich mit dem selben nicht übereinstimme und mich von dem selben

distanzierte, sollten darin Positionen, Haltungen, Bilder etc. enthalten sein, welche

mit dem deutschen Recht nicht vereinbar sein sollten.

Back to top