zyxzyxooxmi

]]>Looks like you, contrary to Suvy, *actually* read Taleb’s work.

What has gone with the development of economics as a science? Answer: There was a bunch of intelligent people who felt compelled to use mathematics just to tell themselves that they were rigorous in their thinking, that theirs was a science. Someone in a great rush decided to introduce mathematical modeling techniques (culprits: Leon Walras, Gerard Debreu, Paul Samuelson) without considering the fact that either the class of mathematics they were using was too restrictive for the class of problems they were dealing with, or that perhaps they should be aware that the precision of the language of mathematics could lead people to believe that they had solutions when in fact they had none. . . . Indeed the mathematics they dealt with did not work in the real world, possibly because we needed richer classes of processes “” and they refused to accept the fact that no mathematics at all was probably better.

— Fooled by Randomness

He is not talking about mathmatical logic; he is talking about mathematical models. He is talking about MV=QP and Y=C+I+G+X-M and econometrics. His particular bete-noir is VaR, Monte Carlo Simulation of returns of a portfolio of assets.

]]>… I wish you luck.

]]>*“That doesn’t imply use of mathematical models.”*

*You have to understand that mathematical models work in certain situations. They break down in certain areas so we can measure how they break down and stay convex to their breakdowns. That way, you don’t need to use math to predict the future.*

Mathematical models do not work in economics because they necessarily presume constancy in a field of inquiry that is not grounded on constants. Knowledge and actions are not able to be regarded as past causally determined on the basis of constants.

*“If you’re making a logically structured argument, because your mind is necessarily only able to think in terms of logic (non-contradiction, identity, excluded middle, etc), then you are logically structured.”*

*“Mandelbrot was a mathematician more closely related to my position, because of his work on chaos and fractals.”*

*I’ve read almost every single paper of Mandelbrot’s work. I still have his books (and his textbook) on my nightstand. I actually make a lot of references to a lot of his works on this blog to support my points. My points are almost the exact same as his on almost every single position. I literally copied his positions because I was so compelled by his work.*

Unfortunately, you are fallaciously imposing mathematical concepts onto economics where they don’t apply, such as constancy in mathematical equations.

*“Monte Carlo simulations predict outcome distribution probabilities which give the researcher the very false sense of certainty that Taleb warns against. Now you’re telling me they should be used more. And you claim to understand Taleb’s main lesson?”*

*The use of math isn’t just for prediction, it’s for the qualitative observation of a system.*

My argument is that observations in economics are necessarily historical. Mathematical modelling in predictions, if they utilize constants, is unwarranted. The constant used in the MBS pricing model for example blew up in the faces of those who utilized it.

*Monte Carlo simulations are very good for this. Not only that, but they are an extremely robust tool that can be used very well for analyzing data. For example, a non-parametric bootstrap is a very useful tool to measure data that is much more robust than other typical statistical tools.*

Not if the phenomena is not constant. Bootstrapping is good for phenomena in which the probabilities are presumed as constant over time.

*“Austrianism is wertfrei. You don’t need to introduce politics. Economics is an a priori science akin to mathematics (not mathematics). It is not an empirical science.”*

*Economics is not akin to mathematics. Starting with a priori assumptions and deducing from them gives very limited results.*

Austrian economics is akin to mathematics in the sense that it is a priori. And nobody claimed the a priori propositions and deductions were *not* “limited.” Yes, it is limited, but all particular sciences are limited.

*“Measurement is necessarily historical. Historical economic information is unique. The data of economic history is determined by people’s knowledge and preferences at the time the actions are made. Humans learn and because of that, history does not tell us what humans are necessarily constrained to do in the future, and there are certainly no constants that are revealed by history that determine future actions.”*

*Not true, take a look at the Taleb paper I posted above on bank stress tests. He advocates using the mathematical model and then testing the model against areas where it fails and staying convex to the model’s failure.*

Not a rebuttal. Stress tests depend on facts that are held as constant. And the stress tests are constrained by the logical categories of action anyway.

*Measuring where the model fails is not a measurement of history; it’s a measure of the limitations of the model.*

A model that utilizes actual empirical values such as bank assets.

*“They do very much, because they are showing you that mathematical models are necessarily unable to describe human action. They show all heretofore models are flawed.”*

*Not true. Just because mathematical models can’t be used to describe human action in a deterministic way doesn’t mean that they’re useless.*

This is also not a rebuttal to what I said. I said mathematical models are flawed if they presume constancy and if they are used in *economics*. Economic history is not economics.

*“Why?”*

*I already said why. Human action is a small part of the economy; there are many, many other factors in the economy as well.*

Everything that is economic is grounded in human action. Every natural phenomena, if it is going to be relevant to economics, is related to action.

*“How do I know he’s not fooled about being fooled?”*

*Read the story about Nero Tulip and John the high yield trader. Read the part where he talks about George Soros. There are other parts where he talks about it in the book.*

Not a response. I am asking you how do I know he isn’t being fooled concerning THOSE VERY arguments.

*“They do very much, because they are showing you that mathematical models are necessarily unable to describe human action. They show all heretofore models are flawed.”*

*When did I say models aren’t flawed?*

When did I say you said they are not flawed?

*You’re putting words in my mouth. Just because models are flawed doesn’t mean they’re useless. There’s a very important distinction here.*

Just because a model is useful for economic history, it doesn’t mean it is useful for economic science.

*I never said that we should only listen to models. I said that models are a tool in the toolkit. They do fail and we have to stay convex to the areas that they fail.*

A tool for history only.

]]>You have to understand that mathematical models work in certain situations. They break down in certain areas so we can measure how they break down and stay convex to their breakdowns. That way, you don’t need to use math to predict the future.

“If you’re making a logically structured argument, because your mind is necessarily only able to think in terms of logic (non-contradiction, identity, excluded middle, etc), then you are logically structured.”

“Mandelbrot was a mathematician more closely related to my position, because of his work on chaos and fractals.”

I’ve read almost every single paper of Mandelbrot’s work. I still have his books (and his textbook) on my nightstand. I actually make a lot of references to a lot of his works on this blog to support my points. My points are almost the exact same as his on almost every single position. I literally copied his positions because I was so compelled by his work.

“Monte Carlo simulations predict outcome distribution probabilities which give the researcher the very false sense of certainty that Taleb warns against. Now you’re telling me they should be used more. And you claim to understand Taleb’s main lesson?”

The use of math isn’t just for prediction, it’s for the qualitative observation of a system. Monte Carlo simulations are very good for this. Not only that, but they are an extremely robust tool that can be used very well for analyzing data. For example, a non-parametric bootstrap is a very useful tool to measure data that is much more robust than other typical statistical tools.

“Austrianism is wertfrei. You don’t need to introduce politics. Economics is an a priori science akin to mathematics (not mathematics). It is not an empirical science.”

Economics is not akin to mathematics. Starting with a priori assumptions and deducing from them gives very limited results.

“Measurement is necessarily historical. Historical economic information is unique. The data of economic history is determined by people’s knowledge and preferences at the time the actions are made. Humans learn and because of that, history does not tell us what humans are necessarily constrained to do in the future, and there are certainly no constants that are revealed by history that determine future actions.”

Not true, take a look at the Taleb paper I posted above on bank stress tests. He advocates using the mathematical model and then testing the model against areas where it fails and staying convex to the model’s failure. Measuring where the model fails is not a measurement of history; it’s a measure of the limitations of the model.

“They do very much, because they are showing you that mathematical models are necessarily unable to describe human action. They show all heretofore models are flawed.”

Not true. Just because mathematical models can’t be used to describe human action in a deterministic way doesn’t mean that they’re useless.

“Why?”

I already said why. Human action is a small part of the economy; there are many, many other factors in the economy as well.

“How do I know he’s not fooled about being fooled?”

Read the story about Nero Tulip and John the high yield trader. Read the part where he talks about George Soros. There are other parts where he talks about it in the book.

“They do very much, because they are showing you that mathematical models are necessarily unable to describe human action. They show all heretofore models are flawed.”

When did I say models aren’t flawed? You’re putting words in my mouth. Just because models are flawed doesn’t mean they’re useless. There’s a very important distinction here.

I never said that we should only listen to models. I said that models are a tool in the toolkit. They do fail and we have to stay convex to the areas that they fail.

]]>*They argue against the predictive power of mathematics in the way it’s used.*

*“The ONLY way it can be used.”*

*THAT IS COMPLETELY FALSE.*

NO IT IS COMPLETELY 100% TRUE.

*It’s much easier and much more feasible to measure sensitivity to rare events than to predict.*

That doesn’t imply use of mathematical models.

*There’s actually a branch of mathematics for it called sensitivity analysis.*

Irrelevant.

*“You just contradicted yourself. In order to even make your argument, you are implementing logic. Telling me “that’s NOT true” presupposes logic of non-contradiction. It is not true, instead of true. It cannot be both true and false. Telling me that we’re not logically structured, is a logically structured argument.”*

*I’m making logically structured arguments, yes. However, humans are not logically structured.*

If you’re making a logically structured argument, because your mind is necessarily only able to think in terms of logic (non-contradiction, identity, excluded middle, etc), *then you are logically structured.*

*We actually have a very poor understanding of living in “fat-tailed” world.*

It is not necessary to my argument that we understand *everything* about the world, including what we perceive to be statistically or probabilistically unlikely events. Even that presupposes logical structure.

*To say that I do everything logically is just wrong.*

It’s clear you don’t understand what logic means. It doesn’t mean you are homo economicus, perfectly weighing all alternatives and doing mathematical calculations to calculate every decision you make, such as where to aim in the toilet, or what friends to make. The logical structure I am referring to is NOT what you seem to assume it is, which is a machine-like, robotic-like calculation machine.

*I’m not logical when it comes to things like friends, family, etc.; I’m emotional.*

You are logical there because you understand your friends according to the basic laws of thought. If you are emotionally attached to someone, you have an understanding of who they are. They are kind, nice, generous, funny, whatever, and these understandings are constrained by the laws of thought. If you understand one of your friends as generous, then you necessarily understand them to be not *not* generous, and so on. A friend cannot be both A and non-A. If they are A, then you logically understand them to be not non-A.

*“Then how is he able to not be fooled? How is he able to know he is being fooled, if he wasn’t certain about his being fooled, which means he can’t be fooled about being fooled?”*

*He actually dedicates several chapters to this in his book; he is fooled.*

How do I know he’s not fooled about being fooled?

*He just understands that he can be fooled. He even gives many examples of this. You should read his work before you tell me about it.*

I have read his work. You just have a different understanding (misunderstanding actually) of his work, and you assume that because I have a different understanding than you, I must not have read his work.

His examples of being fooled presupposes he understands what it is to not be fooled in those particular scenarios. One cannot claim that one can be fooled in X, unless one also understands how not to be fooled in X.

*“I never said they are against using math in economics. I said their arguments support my position, and cast doubt on yours.”*

*They don’t cast a doubt on my position at all.*

They do very much, because they are showing you that mathematical models are necessarily unable to describe human action. They show all heretofore models are flawed.

*My position on this issue actually came from reading their work and theories extensively. Their arguments completely changed the way that I viewed the world.*

Then you haven’t understood what they wrote.

*I actually agree that mathematics in economics shouldn’t really be used the way it is. It should be used for measurement.*

Measurement is necessarily historical. Historical economic information is unique. The data of economic history is determined by people’s knowledge and preferences at the time the actions are made. Humans learn and because of that, history does not tell us what humans are necessarily constrained to do in the future, and there are certainly no constants that are revealed by history that determine future actions.

*It should be used to understand the qualitative behavior of the system. Ironically enough, I don’t think mathematics should be used for prediction in the way that it is. Mathematics is best used for sensitivity analysis. Monte Carlo simulation should be used much more often than it.*

Monte Carlo simulations predict outcome distribution probabilities which give the researcher the very false sense of certainty that Taleb warns against. Now you’re telling me they should be used more. And you claim to understand Taleb’s main lesson?

*Chaos Theory and Nonlinear Dynamics should be used much more. Traditional statistical methods should be used much less since they fail when dealing with rare events.*

So do Monte Carlo simulations.

*“I have. I suggest that you read Austrian economics before you talk about it. It is obvious you have not read a single Austrian text on method.”*

*You haven’t even read his Fooled by Randomness.*

Yes, I have read Fooled by Randomness. You haven’t read Austrian works.

*As for Austrian works, the first book I read in economics was The Road to Serfdom. I read some other papers and got really turned off because they seemed too ideological and these people didn’t seem practical. I tried to read Rothbard’s A Monetary History of the United States and I just couldn’t make myself do it.*

So you haven’t read Austrian methodology.

*By the way, I actually am a libertarian. I just don’t want to read about politics and beliefs when dealing with a science.*

Austrianism is wertfrei. You don’t need to introduce politics. Economics is an a priori science akin to mathematics (not mathematics). It is not an empirical science.

*As for the article you posted, there’s one thing I disagree with.*

*“Economics, on the other hand, starts from an axiom that is known and meaningful to us “” human action.”*

Why?

*Again, I actually agree the most with the Post-Keynesian approach to economics.*

Post Keynesianism is chock full of economic contradictions.

]]>I misspoke earlier when I said Rothbard’s A Monetary History of the United States

I meant his book A History of Money and Banking in the United States.

]]>THAT IS COMPLETELY FALSE. It’s much easier and much more feasible to measure sensitivity to rare events than to predict. There’s actually a branch of mathematics for it called sensitivity analysis.

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

“You just contradicted yourself. In order to even make your argument, you are implementing logic. Telling me “that’s NOT true” presupposes logic of non-contradiction. It is not true, instead of true. It cannot be both true and false. Telling me that we’re not logically structured, is a logically structured argument.”

I’m making logically structured arguments, yes. However, humans are not logically structured. We actually have a very poor understanding of living in “fat-tailed” world. To say that I do everything logically is just wrong. I’m not logical when it comes to things like friends, family, etc.; I’m emotional.

“Then how is he able to not be fooled? How is he able to know he is being fooled, if he wasn’t certain about his being fooled, which means he can’t be fooled about being fooled?”

He actually dedicates several chapters to this in his book; he is fooled. He just understands that he can be fooled. He even gives many examples of this. You should read his work before you tell me about it.

“I never said they are against using math in economics. I said their arguments support my position, and cast doubt on yours.”

They don’t cast a doubt on my position at all. In fact, my position is almost the exact same as theirs; I’m being dead serious on this. My position on this issue actually came from reading their work and theories extensively. Their arguments completely changed the way that I viewed the world.

Here’s a quote by Taleb on this point.

“Tragically, before the proliferation of empirically blind idiot savants, interesting work had been begun by true thinkers, the likes of J. M. Keynes, Friedrich Hayek, and the great Benoit Mandelbrot.”

I actually agree that mathematics in economics shouldn’t really be used the way it is. It should be used for measurement. It should be used to understand the qualitative behavior of the system. Ironically enough, I don’t think mathematics should be used for prediction in the way that it is. Mathematics is best used for sensitivity analysis. Monte Carlo simulation should be used much more often than it. Chaos Theory and Nonlinear Dynamics should be used much more. Traditional statistical methods should be used much less since they fail when dealing with rare events.

“I have. I suggest that you read Austrian economics before you talk about it. It is obvious you have not read a single Austrian text on method.”

You haven’t even read his Fooled by Randomness. As for Austrian works, the first book I read in economics was The Road to Serfdom. I read some other papers and got really turned off because they seemed too ideological and these people didn’t seem practical. I tried to read Rothbard’s A Monetary History of the United States and I just couldn’t make myself do it.

By the way, I actually am a libertarian. I just don’t want to read about politics and beliefs when dealing with a science.

As for the article you posted, there’s one thing I disagree with.

“Economics, on the other hand, starts from an axiom that is known and meaningful to us “” human action.”

Again, I actually agree the most with the Post-Keynesian approach to economics. It seems to me that the Post-Keynesian approach is closest to what Keynes actually argued that economics should be about. I don’t agree with the neoclassical approach to economics at all.

]]>A Note On Mathematical Economics

For a realization that mathematical logic is essentially subordinate to the verbal, see Vocabulaire Technique et Critique de la Philosophie, pgs 574, 579. Hope you can read French.

]]>