Talk:Model (economics)/Archive 1

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

How about

In economics, the term model denotes a theoretical construct that represents economic processes by a set of variables and logical and quantitative relationships between them.

I think this works better as a first sentence than mine because it defines what a model is, whereas mine discussed what they are for. I like the idea of something like your sentence first and then my first sentence second. The more I read your sentence above, the more I am convinced of its accuracy. However, as it is now, I think it would make it very difficult for someone that does not already know what a model is to understand it. Maybe it is just the order of the sentence, where the 'and and' contruction makes it less approachable.

If this representation is based on empirical tests that determine the validity of the model within a specified range, we will say the model is an empirically justified model.

This is a needed concept in the article, but not necessarily needed in the first paragraph.

Models are not expected to provide fully accurate acccounts of economic processes.

That is essentially what I was trying to get at, but I also felt my wording showed why models are not accurate representations of reality: they carry less information by design, in order to simplify.

from here on below, go ahead, I have very little knowledge of economic history or marxist theories.

also need

History. As far as formal models is concerned, what can we say about originators? I would guess one of

William Stanley Jevons

Leon Walras

but I dunno. What about David Ricardo? Any ideas?

Also Marxian models of Surplus value would be interesting to say something about.

Are Accounting models tautologies?

I'm not sure I understand what you mean. How would they be?

This is an important page! CSTAR 20:32, 6 May 2004 (UTC)

Thank you. I thought so. Also thanks for collaborating and your insightful additions. Its always a little hard to write something you thought was pretty good and to throw it out there for others to point out rightfully where it is not. :) - Taxman 23:29, May 6, 2004 (UTC)

Accounting Models

I think of an accounting model (such as those in National Income Accounts) as essentially being represented by

(a) A graph of flows

(b) At each node some identity expressing that

       algebraic sum of inflows = sink

(e.g. one of Kirchoff's laws) Thus in the Keynesian model

Y = C + I + G + X - Im is a tautology

C = C(Y) is not.

History

I suggested Marxian, not because I think it is particularly accurate but as a balance to more academically (at least in US) polite models. CSTAR 23:43, 6 May 2004 (UTC)

More work

probably needs to be done on this page. Feel free to change or remove anything I wrote (that goes without saying in a wiki I guess)

I wanted to discuss Samuelson's concept of operationally meaningful theorem and will do so soon. As I mentioned earlier, I would also like to see whether earlier economic models in the 17 and 18th centurry existed. This sound like a very French idea, so maybe some predecessor of Leon Walras (for example Francois Quesnay) had some models. Talking about econmics per-se (a la Adam Smith, or Hobbes or Hume) does not a model make. I am not a historian though some my knowledge is very mundane, limited to what I know from Heilbronner and a few other popular references.

Again if you have any ideas CSTAR 03:33, 7 May 2004 (UTC)

Historical precedents

Yes it seems that the French physiocrats really did have discernible models.

All through the 18th century simple probabilistic models were used to understand the economics of insurance. This was a natural extrapolation of the theory of gambling, and played an important role both in the development of probability theory itself and in the development of actuarial science. Many of the giants of 18th century mathematical physics contributed to this field. Around 1730, De Moivre addressed some of these problems in the 3rd edition of the Doctrine of Chances. Even earlier (1709), Nicolas Bernoulli studies problems related to savings and interest in the Ars Conjectandi. In 1730, Daniel Bernoulli studied "moral probability" in his Mensura Sortis, where he introduced what we would today call "logarithmic utility of money" and applied it to gambling and insurance problems, including a solution of the paradoxical Saint Petersburg problem. All of these developments were summarized by Laplace in his Analytical Theory of Probability (1812). Clearly, by the time David Ricardo came along he had a lot of well-established math to draw from. — Miguel 06:17, 2004 May 10 (UTC)

Miguel, que tal si agregas esto, en la parte de historia? CSTAR 03:36, 3 Jun 2004 (UTC)

Hecho. Miguel 20:11, 3 Jun 2004 (UTC)

Featured article ready?

Due to many great edits by CSTAR and Miguel, I think this article is ready to be a featured article candidate. It will get some good criticism or be confirmed as a featured article. I know the knee-jerk reaction would be that the article needs a picture or a diagram. So I put in a diagram of the IS/LM model, which will at least act as an example. Let me know what you think of using that diagram and the featured article candidacy. - Taxman 18:40, Jun 24, 2004 (UTC)

We also need to say something about econometric models.CSTAR 21:43, 24 Jun 2004 (UTC)

ok, well get after it then! :) I'll try to put some basics together at least. - Taxman 00:30, Jun 25, 2004 (UTC)

Yeah I intend to shortly.CSTAR 00:32, 25 Jun 2004 (UTC)

OK I added something, I may be too high fallutin' (hey I'm a mathematical physicist or something), so have a look. Covariance invariant is not a standard term, but I can define it. For instance, Black-Scholes uses more than the covariance structure. Any way also need references aand I'm at home and won't go to the library till next week.CSTAR 01:14, 25 Jun 2004 (UTC)

Need to coordinate

OK we both added stuff; you can integrate it somehow with my addition. Delete whatever you want.CSTAR 02:07, 25 Jun 2004 (UTC)

If necessary, I'll coordinate, but please say what you want to do.CSTAR 03:45, 25 Jun 2004 (UTC)

BTW I like the picture.CSTAR 04:44, 25 Jun 2004 (UTC)

Operations Research

I'd like to relate economic models to operations research, but I can't find a way to integrate that with the article. help? Miguel 05:01, 25 Jun 2004 (UTC)

I don't think it is correct o say that a model ""more generally achieves some goal. Model as defined is about as general as one can get. If anything one should say more specifically or perhaps, in general termsCSTAR 13:40, 25 Jun 2004 (UTC)

I made some more edits. I moved around some stuff, because I wanted to keep the methodological and structural differences clear. These different classifications I think are useful, particularly the conventional model concept, which is an important thing to keep in mind in present day economic discussions (in US at least) CSTAR 20:36, 25 Jun 2004 (UTC)

Unfortunately I think that distinction has come at a major cost to clarity and readability. You may know what you mean by structural, abstractions, stochastic, fine grained, covariance invariant, and the methodological distinction, etc., but the article doesn't. Even linking to those concepts would not be enough, they would need to be defined in stream. I believe a wikipedia article needs to explain every term and concept that would not be known to a person that doesn't already have a familiarity with the subject. In addition, after reading the article, I still don't see what value the distinction has. Even then, if the cost to readability is so great, it could oversahdow the value of the distinction for 99.9% of the audience the article would otherwise be valuable for. - Taxman 21:21, Jun 25, 2004 (UTC)

OK we can get rid of that sentence on models of finance. Agreed it's not very meaningful.

It has value, just maybe not in this article. perhaps as a comment in the general model article.

But the distinction between statistical and non-statistical is too important to leave out; moreover, the concepts of validity by convention and empirical validity (perhaps using other phraseology) is also important.

certainly. I was not debating these. That last distinction is pretty clearly written to me.

For instance, much of the discussions on SS can be separated neatly into two parts: The institutional structure of social security and the financial structure (e.g. present value of various streams of income).

I have no idea how that applies to economic models, or at least to the study of their general features.

Admittedly, the article has to be clear, but it also has to say something useful and interesting that will contribute to understanding of other economic issues.

You have contributed much that already is useful and interesting. I'm just not sure some of the latest edits offer greater value than their cost. The distinction I was most worried about is the distinction you referred to in your comment above: methodological vs. structural. That is where I am not sure the value of the distinction pays its cost in readability and clarity.

At any rate, please go ahead and edit it. CSTAR 21:57, 25 Jun 2004 (UTC)

Part of what I was getting at is that I don't know what you're trying to get accross with some of what you've written either. Therefore I can't properly edit much of it. Until you flesh out, or you chose to remove some of those ideas I noted above, I don't feel there is much I can do. I don't want to ever remove valuable material, its just that I don't see where some of it fits in yet. Part of that is I am not at your level of understanding on the issue. But that is where I think I can be most valuable in offering comments on the readability to people at my level and below (>99.9% of the article's potential readers, even simply considering how many english speakers have a degree in economics and mathematics). So all my comments are of course just an attempt to get towards a better article, not a personal attack. - Taxman 22:41, Jun 25, 2004 (UTC)

Thanks. All right, I'll try to see what I can do.CSTAR 22:44, 25 Jun 2004 (UTC)

Latest version

Please have a lookCSTAR 23:19, 25 Jun 2004 (UTC)

I have recently discovered this article and was very impressed. The constructive criticism I would like to make however, is that the basics need to be filled in. This reads more like an intermediate textbook than an encyclopedia article. Don't misunderstand me: I am not saying the article should be "dumbed down". There is no reason we cannot write to an advanced undergrad audience or even higher. However, before we use any advanced concepts we have to cover the basics. A well written article builds from the simple to the more complex. To this end I added some stuff on the simplification and abstraction roles of models and a bit on the pitfalls of improper model construction. mydogategodshat 02:23, 26 Jun 2004 (UTC)

I took your comments at the beginning and rephrased them in a form which I thought was faithful to what you wanted to say. Two sentences I removed, because I thought they were included in other comments later on. Any additions of course will be welcomed by all of us who worked on it. I don't want to seem possesive, but I think aggressive editing on may be necessary to produce something. Of course that works in all directions.CSTAR 04:12, 26 Jun 2004 (UTC)

PS Do you still think more introductory material is necessary?CSTAR 04:37, 26 Jun 2004 (UTC)

Something seems to be wrong with the indentation structure of the "types of models" section.CSTAR 18:11, 27 Jun 2004 (UTC)

What is wrong? Can you be more specific?mydogategodshat 18:26, 27 Jun 2004 (UTC)

The indentation immediately after the displayed equation isn't the same as the indentation immediately before the displayed equation.CSTAR 18:37, 27 Jun 2004 (UTC)

PS. For instance, in the text:

expresses some principle of conservation in the form

algebraic sum of inflows = sinks - sources

This principle is certainly true for money and it is the basis for national income accounting. Accounting models are true by convention, that is any experimental failure to confirm them, would be attributed to fraud, arithmetic error or an extraneous injection (or destruction) of cash which we would interpret as showing the experiment was conducted improperly

I'm not sure exactly what the problems is but in my browser (opera) the text expresses some principle of conservation in the form is further to the left than This principle is certainly true for moneyCSTAR 18:40, 27 Jun 2004 (UTC)

Forget it, I see this other places as well (in other articles).CSTAR 18:45, 27 Jun 2004 (UTC)

I hope that the two level categorization scheme was what you intended. If not, change it to a single level with four items. It seemed to me that the basis of categorization of the first two items was significantly different from the basis of categorization of the latter two items, so I gave it two levels. I hope that doesnt change your intended meaning. I do fell it is very important for this section to have some sort of structure: It make the material much clearer (at least for me). mydogategodshat 18:58, 27 Jun 2004 (UTC)

No no, I don't object to that. I just found at odd that there appeared to be a 3 level categorization, but this is probably a bug of some kind, somewhere.CSTAR 19:07, 27 Jun 2004 (UTC)

I exanded on something mydogategodshat wrote earlier on On a more practical level .... I believe what I wrote is still true and useful and correctly expands the intended meaning of the original sentence.CSTAR 00:03, 28 Jun 2004 (UTC)

In fact, I think one more useful item could be added to that structured list: In addition to accounting and optimizatio model, we should also say something about aggregation model. By that I mean a macro model obtained by aggregating quantities from a micro-level. Thus in Macro models, consumption as a function of income is an aggregation of individual consumption decisons. Now whether we write C=C(Y) a la Keynes or something else (permanent income hypothesis) depends on phenomenological considerations or some micro principle. CSTAR 00:14, 28 Jun 2004 (UTC)

Your distinction between optimization models and aggregation models sounds alot like the common distinction between partial equilibrium models and general equalibrium models. Should this be included? What about dynamic models versus comparative static models? How about matrix models like input-output tables, game theory, and Physiocratic models? Where do these fit in your categorization scheme? Should we say anything about complexity models, chaos theory models, and behavioural economics modeling?mydogategodshat 03:09, 28 Jun 2004 (UTC)

I'm try to remember if there are any dissaggregation models in economics. There are several in management (for example: advertising budget estimation and top-down sales revenue calculations). The closest I can think of are duopoly and ologopoly models that start with industry data and work back to results specific for a firm given assumptions about interactivity etc. mydogategodshat

Yikes. I mentioned something about input-output models. I guess that I sort of gave up trying to set up a general taxonomy of models. Duopoloy and oligoploy (as treated for example in Henderson and Quandt) could probably fit into optimization and comparative statics. Nothing is mentioned on dynamic models as opposed to comparative statics. Although we could inclde business cycle models.CSTAR 04:11, 28 Jun 2004 (UTC)

Comments on new intro sentences

I have several objections to the recent changes in the lead paragraph, summarized thus:

Models don't have to be simple; the lead paragraph doesn't say that exactly, but the contrast simple/complex emphasized in the current wording is not art of the concept of model in general, in my opinion. Some models are indeed simple, and if they're too simple one usually says they're caricatures (which is not to say that some of these caricatures can be useful conceptual and expository tools, however). The fact is that many models are extremely elaborate, for instance models constructed in the Jan Tinbergen tradition which tend to be computationally very intensive.
Models may be constructed for many reasons. Formal mathematical models by themselves probably don't do much "illumination"; their usefulness results from the analytical tools they provide to economists. Of course you may think I'm too being much too picky about the selection of words. But why does the use of "illuminate" improve what was there before?

CSTAR 01:13, 27 Dec 2004 (UTC)

1. Yes, but models are always simpler than the reality that they are trying to approximate. That is an important distinction and part of the reason why they are used. Models can be extraordinarily complicated, but that doesn't change the fact that they are simpler. 2. The changes make the text easier to understand for someone that doesn't already know what the subject means. I believe that is important. - Taxman 21:14, Dec 27, 2004 (UTC)

Intro needs to be simple and to the point. Maybe Taxman's point (i.e. models are like maps - they simplify reality in (hopefully) useful ways; if they represented the whole of reality, they would serve no function) could be worked in without over-complicating it. Rd232 01:24, 28 Dec 2004 (UTC)

Point taken. Please understand that my criticism is meant in a friendly way, of course: I believe that every contributor to this article has not had any other agenda than the attainment of absolute clarity (which includes myself, although I may have not consistently achieved that goal).

CSTAR 21:45, 27 Dec 2004 (UTC)

Model (macroeconomics)

That article seems woefully unaware of this one. Perhaps it needs to be merged in here? - Taxman 16:30, Mar 31, 2005 (UTC)

Yes--CSTAR 16:41, 31 Mar 2005 (UTC)

Is any of it accurate though? I left a question regarding its source on the talk page, hopefully the author of the material is available. - Taxman 17:41, Mar 31, 2005 (UTC)

Hubris and other opinions

The section on hubris is a bit odd. I have no real objection to the main point, (which seems to be a comment on the ineffectiveness of central planning) but it could be made more directly based on what already is discussed elsewhere in the article. Also is "affect" a spelling error or are there some aspect of models of which I am not familiar?

The reference to chaos in the previous paragraph is someone's personal opinion. Please, note that meteorology is based on fluid mechanics and has basic physical principles supporting it, such as Navier Stokes and application of Coriolis forces. Economics is nowhere near the stage yet, that is where we can say "it's in principle this". You're knocking down a straw man. --CSTAR 17:21, 20 May 2005 (UTC)

Hi, neither my spelling nor my reasoning are very careful in a first draft I'm afraid. I expect readers like you to call attention to such things, it you are reluctant to fix them yourself. (Please feel free if you have the time). I will remove the chaos claim myself if I am unable to find an economic model clearly subject to small-perturbation butterfly-effects. (If I can't I'll have only a hypothesized a strawman for the future rather than attacked him - but still).

Thanks for reading, Wragge 18:06, 2005 May 20 (UTC)

I am also worried that much of what has been added to the article suffers from being opinion instead of fact. It reads really well, so don't get me wrong, I certainly appreciate the contribution, but for a great encyclopedia article there should be more. A closer adherance to the Wikipedia:Verifiability policy and keeping in mind the Wikipedia:No original research policy would help significantly. I hope this doesn't come accross as abrasive as I am worried it might be, it certainly is not intended to. - Taxman 18:14, May 20, 2005 (UTC)

Non-well posedness of initial value problems

The kind of behaviour (such as the "butterfly effect") you refer to as "chaos" has a precise technical name, namely non-well posedness. Consider a problem whose temporal behavior is described by a function Φ of two variables: an initial value and time.

$\Phi(x_0, 0) = x_0 \quad$

Non-well posedness in this instance means that for values of t>0, the mapping

$x \rightarrow \Phi(x, t) \quad$

is not continuous (or fails to have other differentiability properties)

Chaotic behaviour refers to very specific kind of initial value problems which has attracting sets (this expresses the limiting behavior in time) which are fractals, e.g. sets with self-similar behavior usually identified with sets of fractional Hausdorff dimension. This behavior may or may not be associated with non-well posedness.

It's not clear what you're trying to achieve by these edits; I would suggest you discuss your goals before going any further. Unfortunately, I realize I sound abrasive with these comments, but the article seems to ramble on with various editorial opinions on models, which though potentially useful, need to be stated in much more succinct form. --CSTAR 21:07, 20 May 2005 (UTC)

Here is my reply to the concerns raised above:

There appears to be a vast literature on chaos in economic models, so I suppose what I was "trying to achieve" was to add to the "Economic model" page an aspect of economic modelling not already mentioned. (This is an encyclopedia - which I take to mean covering everything).

Do you think it would be better to add this to the chaos page? Surely, it should at least be mentioned here, as it is still a very hot topic in the field. Possibly I am misunderstanding the intent of this page, is it intended only to be a brief overview?

I absolutely agree with your point on concision, please compress and combine the points as much as possible (that's what I try to do).

Wragge 21:24, 2005 May 20 (UTC)

By the way, "non-well posedness" is an excellent name, which I think should be included, but this shouldn't preclude the more widely-known term "butterfly-effect" should it?

Wragge 21:29, 2005 May 20 (UTC)

Is this article intended as a general discussion of economic models?

I had assumed that this article was aiming to talk about the history of "modeling" in economics - but perhaps this is where I went wrong?

Alternatively there may be something else about this page which I am not aware of; is it a candidate for article of the day? Is it part of a larger plan? I don't believe in plans (see points raised in the article) - I just type when subjects seem to be missing (I assume that if I add a detail in the wrong place it can be moved later).

Nothing much different with this page than others. There are just two experienced Wikipedians, myself and CSTAR that have been working on this page for a long time. We want to make it the best article we can, as I'm sure you do, but we appear to have a better handle on Wikipedia's policies. Nothing wrong with that, just keep trying to help and learn how to help the best. - Taxman 04:34, May 22, 2005 (UTC)

I had a look through the discussions above and detect some of the same confusion. If I knew more clearly what the article was trying to focus on I might be able to contribute better.

The Wikipedians here seem to be rigorous and intelligent, so I'm sure you will appreciate the irony of fate which has befallen me: I have been more severly challenged on this page than on any of my other contributions even though this should be the one page I actually know something about (in a minor way, this has been my job and educational background for the past ten years).

Of course, the other irony is that this page appears to have a tighter control system than any other (where one may have expected a more free-wheeling laissez-faire attitude).

I will wait until the issue of what this page is aiming to do is cleared up, before making any changes (which prevents me from citing sources as requested by the User:Taxman).

Well the standard Wiki way is that an article on a direct topic, like this one, should cover the topic in a broad manner, cover every important facet of the topic, and not cover any one topic more than it's importance to the subject deserves. That's the ideal anyway. Yes that means this is the overview article. As such it contains some history, but should not focus on it. A 'History of economic modelling' article would be the place to cover that as its topic.

Strange that you would say a question of what the topic of this article is prevents you from citing sources. It does no such thing, and is entirely mutually exclusive. Whatever the topic is, be it snails or economics, all material added to Wikipedia can be cited. Cited material that is off topic can be moved, uncited material is much harder to work with later. - Taxman 04:34, May 22, 2005 (UTC)

By the way, I like an abrasive approach - that's the proper way to express a difference of views - Cheers Wragge 22:20, 2005 May 20 (UTC)

Interesting approach. It certainly works at times, though I prefer to use influence and requests instead. In any case, your efforts are certainly welcomed, as long as you make an effort to follow Wikipedia's policies. I decided to just respond here before delving into the volumes of material below. - Taxman 04:34, May 22, 2005 (UTC)

Hi Taxman, I can understand how that would have looked strange - there was a three way conversation going on: first of all you asked me to cite sources, and I was half way into doing this (adding them to the chaos bit - although probably too many now) before another user (CSTAR) asked me to stop completely. My reason for stopping was simply that he asked me to dissist, and had nothing to do with your earlier request for sources. I only mentioned your name as an apologia for not fulfilling the citation request.

This article seems to include a history of economic modelling already, so a new article on the history would presumably be based on this if one were ever made. Thanks for asking for more details, but I don't think I'll add any more to this article, because: (1) I don't really think my tactics are compatable here, (2) I don't have a compelling need to write on this topic - or even much interest in it as a theoretical overview, (3) There are 500,000 other articles for me to 'improve'.

Cheers, Wragge 05:16, 2005 May 22 (UTC)

I understood the three way conversation. The approach I was referring to is what you stated "I like an abrasive approach". Even more interesting for someone that prefers that approach is that you immediately stop your efforts when challenged. If you have something to offer the article please do, just be willing to mold your efforts a bit for greatest effectiveness. If you choose not too, so be it, carry on helping wherever you so choose. I still haven't had time to read all of the changes or even all of the commentary, so for all I know, maybe it would make my life easier of you edited other articles instead of this one! :) - Taxman 05:50, May 22, 2005 (UTC)

Hi Taxman,

I'm glad you were following that. I like an a forthright exchange of ideas, and therefore I definitely do not stop making additions when challenged. As you see in the history I immediately added a section backing up the "chaos" point when you (correctly) challenged me for perhaps introducing a strawman. HOWEVER; I then received a direct request to cease additions. There is a difference between a request like that and a request for more information. I am sure we are all tired of this conversation now, so please don't feel compelled to read the detailed answer below - its mostly about semantics and collaborative philosophy.

Cheers,

Wragge 13:19, 2005 May 22 (UTC)

Wragge : You twice make the claim that I asked you to desist:

and I was half way into doing this (adding them to the chaos bit - although probably too many now) before another user (CSTAR) asked me to stop completely. My reason for stopping was simply that he asked me to dissist,

Please look at what I actually said -- discuss your goals before going further. For what is a collaborative effort, that is certainly a reasonable request; it's a request for discussion and is not a request for you to stop. Notice also that I didn't ask (or even suggest) you delete any of your edits. Moreover, my list of comments below were written asking for clarification of terms after I made an effort to edit your additions. For instance, I asked a question about present utility, because quite honestly, I didn't know which one of the several possibilities I memtioned you were referring to. The same problem occurred in several other instances. --CSTAR 15:53, 22 May 2005 (UTC)

I agree - you've done nothing wrong - and some of my points were genuinely unclear. I think that's an excellent place to end this discussion. Cheers, Wragge 17:39, 2005 May 22 (UTC)

Further comments

I attempted to edit the article, starting with the section Omitted details:

A great danger inherent in the simplification required to fit the entire economy into a model is omitting critical elements

This complexity has been mentioned several times in the article, most notably in the Overview section. It should be clear by now that this is the whole point of models.

Some economists believe that making the model as simple as possible is an artform, but the details left out are often contencious.

Again this has been mentioned before in the article, and by now it should be clear that the point of models is to make them as simple as possible

Market models often ignore externalities such as unpunished pollution.

OK good point, although the question is not punishment, but that of assigning and contractually enforcing payment for costs, such as pollution.

Such models are the basis for many environmental attacks on mainstream economists, for including the price of everything but the value of nothing in their models.

I would have preferred myself the use of a more neutral term than attack. Why do you link to dismal science? This seems excessively gimmicky. Your formulation including the price of everything but the value of nothing seems to me at least innappropriate for an encyclopedia article It also suggest that the environment has value but no price. I believe that's wrong and is an anti-environmentalist POV.

In turn, environmental economics has been accused of ommitting key financial considerations from its models. For example the returns to solar power investments are sometimes modelled without a discount factor, so that the present utility of solar energy delivered in a century's time is precisely equal to gas-power station energy today.

OK environmental economics has indeed been accused of a lot of things, so that it is accused omitting financial details is certainly nothing new. Go ahead and say it, But what is present utility?

Do you mean present value of some future stream of costs and expenses?
Do you mean individual utilities aggregated and discounted like a stream of income?
Do you mean some kind of social welfare function discounted in the same way?

Financial models can be oversimplified by relying on historically unprecented arbitrage-free markets, probably underestimating the chance of crises, and under-pricing or under-planning for risk.

Estimation of statistical parameters from time series data can be innaccurate. I sdmit to not understanding your reference to historically unprecented (sic) arbitrage-free markets. Do yo mean that in the past arbitrage played less of a role than it does now and so therefore time-series data is less reliable? Your use of Kurtosis is very strange. Kurtosis as you know is some normalized form of the 4^th moment

$\operatorname{E}(X - \operatorname{E}(X))^4$

while this does have some bearing on outliers, to identify this value as the chance of a crisis is idiosyncratic.

Models of consumption either assume that humans are immortal or that teenagers plan their life around an optimal retirement supported by the next generation.

Are you making some reference to macroeconomic models such as permanent income hypothesis which try to include life-cycle calculations into the model? This comment is far too vague to be of any use here, and should be in some other article on macroeconmic models.

--CSTAR 16:57, 21 May 2005 (UTC)

Response to CSTAR’s concerns by Wragge

Wow,

CSTAR has made a detailed and interesting analysis of my additions here. I don't completely agree with all of the points raised, but a lot of them are undoubtedly correct and I think they all have merit. Before I attempt to address them specifically (see below) I'll raise a few general things which cover all the subjects here:

In terms of detail-level the two risks to avoid are excessive and repetitious argument on the one hand, and lack of explanation and citations on the other. However, these are both a function of what level the article is pitched at - if it is intended to be a simple introduction then my oblique connection of kurtosis to crisis would certainly be confusing (although reading the link would show how they are mathematically connected).
Repetition occurs throughout Wikipedia, even on the same page; although in theory this is bad it should be remembered that not everyone will read an article from top to bottom. If someone were interested in Model (economics)#history they might simply read that text under that heading, and wouldn't necessarily know any points Model (economics)#overview had made. To account for this I always try to make the text under each heading as standalone as possible (for instance redefining acronyms which have been given earlier). An important advantage of this approach is that if each heading forms a distinct logical "component" it becomes easier to turn it into a separate article when it becomes too long. There are drawbacks of this approach, which is much more relevant to incomplete articles (articles under development) than those which are essentially finalized and ready for featured article status. To be honest with you, I didn't think that this topic was covered comprehensively enough to be one of those.
CSTAR seems to understand the point I was (inchoately) trying to make in referring to such things as the comical assumptions of immortality implicit in some models. Rather than raise my imprecise language here, I would hope that he simply cleans it up, to reach the standard of mathematical precision to which we all aspire? The comment above says that an edit has been attempted, but I see no new versions. Is this edit in another place?
Remember, Wikipedia is a collaboration; if you grant that I've raised some points which were relevant and new, then, rather than pick holes in the language on this talk page I think it would be better to clean up the spelling and scientific terminology in the article itself. In this way, we create a division of labour: I bring up some ideas, you state them more exactly; some-one else adds a critique, a further person adds a counterpoint, and then you come back and clean up the spelling.
The language I've used can too easily be interpreted as having a non-neutral POV; again, this was a first draft, and I assumed I could leave it to others to cleanup any inadvertent appearance of opinion. However, as a matter of fact this is one of the few subjects on which I don't have a strong opinion.
In regard to the points above I have tried to practice what I preach and I actually made some changes to this article cleaning up (as I saw it) some questionable language and generalizations before I added any new points. For instance, an earlier version claimed that because humans are not Homo economicus the conclusions of economic models must be false, I clarified the logic to say that the conclusions would merely be inapplicable (irrelevant).
Model (economics) is a fairly non-specific article (as far as I can tell) it seems as though it would be hard for anyone to have strong opinions about the subject of "modeling in economics" one way or another. The accusations of expressing opinions are probably more about the language I've used (which I admit was pretty careless) than an accusation of bias, although I could be wrong here (am I?)
Some of the points which now appear to be made in sections I've written were actually moved there when I broke the article down by adding more headings; the points about environmental economics (for instance) were simply moved from the earlier version.
I agree that my chaos heading was confusingly detailed (and possibly rambling) but in this case I have a strong defense for the number of citations given in that I had actually been challenged earlier that day (on this page) to provide evidence that economic models and chaos exist.
Although an abrasive critique is appreciated, a dismissive one is not; redefining common terms such as Kurtosis or using expressions such as “rambling” gives the impression that one Wikipedian is implying incomprehension on the part of another (rather than trying to expand comprehension). CSTAR may (or may not) have a greater general understanding of these topics than does the average Wikipedian, but to leverage The Wisdom of Crowds we should encourage rather than (appear to) dismiss those who are less expert than ourselves. Although much of what I have written can be improved on, it should be improved, rather than argued about on a talk page; however knowledgeable any individual Wikipedian is about a subject there are always other points of view which can enrich an article. (The many are smarter than the few, which is the ultimate reason that Wiki-style sites will supplant read-only sites. In the long run it is the reason that Encyclopædia Britannica Inc. has no answer to the Wikipedia challenge.)

Addressing (some) specific criticism of my edits

I don't (and can't) defend every change I have made, but since specific questions have been raised here, I'll give the best specific answers I can:

CSTAR said: This complexity has been mentioned several times in the article, most notably in the Overview section. It should be clear by now that this is the whole point of models.

My defense: To repeat what I say above; repetition is not, in itself, a massive problem (in my view). In order to create a self-contained unit of text under a heading it may be necessary to describe the subject in a similar way to that given in the overview. (Perhaps this is against policy in the long term, where every article is designed to be essentially static and read from beginning to end, but for a page under development I think more headings are almost always better.)

CSTAR said (about simplifying economic models as an artform): Again this has been mentioned before in the article, and by now it should be clear that the point of models is to make them as simple as possible

I have two separate defenses:

Although it has been stated in the overview that the skill of economic modeling is simplifying the economy, it is has not been explained that this was considered an art in itself (many citations available I believe). This is an interesting topic which could be expanded on (Paul Krugman has done so in several of his popular books).
Should I repeat my defense of repetition a third time? Ok, here goes: Even leaving aside the 'modularity' argument repetition is a good thing for a reader who always starts at the top and reads to the bottom of an article because the salient points can be reinforced through repetition. Have you got that? How can salient points be re-enforced? That's right: through repeating them. If you've read this far I bet you aren't going to forget the point I'm trying to make (that's the power of explicit repetition, see: implicit repetition). Admittedly, Wikipedia is not a PowerPoint presentation (in which the reader must first be told what they are going to be told, told it, and then told what they have been told) but equally repetition is a common phenomenon here and (in my opinion) not necessarily a bad thing.

CSTAR kindly said (about Market models often ignore externalities such as unpunished pollution”): OK good point, although the question is not punishment, but that of assigning and contractually enforcing payment for costs, such as pollution.

Defense: Assigning and enforcing costs is "punishment" from the point of view of a profit maximizing firm (especially in a market model trying to incorporate pollution and provide a free market solution). It is the threat or implementation of punishment which is designed to change their behaviour. Although the work "unpunished" appears to be value-loaded, and may imply state-intervention to "punish" through the law, the word does not necessarily hold this meaning for most readers (or in a dictionary definition, see: Wiktionary:punishment). The introduction of this idea to the article was not originally made by me, this was one of the points made before the fateful day I arrived here, the only change I made was to add the word "unpunished", since without this word the line includes models which simply ignore pollution and models (such as carbon trading models) which are specifically designed to include it.

My intention: By adding to the text I was trying to more specifically identify cases in which the real-world economy still indulges in pollution without "assigning and contractually enforcing payment", but where environmental objections say the model should incorporate the harm done. If a carbon tax is imposed on an industry then this objection no longer applies in the same way (it may apply to the tax itself, but would not be made about the model, since even a model focusing only on the bottom line would now include the cost of the pollution – it would not be an externality).

Claim that my correction was, in fact, an improvement: To put it another way, without the word "unpunished" the example of pollution as an externality was simply factually wrong, because some polluting actions do result in a financial penalty, not external to a straightforward accounting analysis (possibly high enough to satisfy some environmentalists).

Counter-question: CSTAR says that "the question is...". This sounds dangerously like nomative economics - although models have been created specifically to advocate particular policies on pollution, the general "philosophy" behind designing a model is (supposed) to be that one asks the model questions rather than designing it to give the answers required. (This is another topic not really spelled out here - is the design of the model independent of the philosophy and politics of the designer? To what extent is it not?)

CSTAR observed: Why do you link to dismal science? This seems excessively gimmicky.

Mea culpa - CSTAR nailed me there.

CSTAR gave me a useful pointer when he said: "I would have preferred myself the use of a more neutral term than ‘attack’".

I’ll take that advice; it’s always worth getting tips on sounding more neutral – one reason for using the word is simply that when you write about theories a lot you tend to get bored with typing “critique”.

CSTAR made a typo of his own when he wrote:

“Your formulation ‘including the price of everything but the value of nothing’ seems to me at least innappropriate (sic) for an encyclopedia article It also suggest that the environment has value but no price. I believe that's wrong and is an anti-environmentalist POV.”

My defense: This formulation is (in my opinion) a traditional and roughly accurate summary of the typical environmental critique on economic models. If a reader believes that this critique shows an anti-environmentalist POV it might be because they have an anti-environmentalist POV themselves. (I don’t believe I have misrepresented the arguments given by many greens - this is what they say - should I provide a link?)

CSTAR asked: "What is present utility? "

In the Ramsey model, etc, the felicity function is defined as instantaneous (present) utility, but all of the ideas CSTAR mentioned could be given as flaws in models advanced by the green movement at one time or another. They are, therefore, all accurate, and it seems unnecessary to specify exactly what the words “present utility” mean – the sense is conveyed without violating any scientific definitions. If we are going to get into greater detail about the specific meaning of this term it will probably be better to create a new article: Critique of Green economic models.

For Ramsey model see: [1]

CSTAR WROTE: "I sdmit to not understanding your reference to 'historically unprecented (sic) arbitrage-free markets'"

My response: On the contrary, I believe that CSTAR does understand exactly what I was trying to get at here; I think his objection is simply that I expressed myself imprecisely and with poor spelling. (I like his elegant formulation "I admit to not understanding…", but feel he undercut this by pointing out an easily-fixable spelling mistake.)

My defense: A consideration here is that the subject of financial modeling requires at least a page to cover adequately, so I had to consider brevity. It is true, that in doing this, I erred and became semantically confused. To spell out some of the points I was trying to encompass in this short line:

The Peso problem and survivorship bias, etc, systematically cause an underestimation of volatility parameters in econometric regressions.
By attempting to fit historical data into a market model with normally distributed shocks the act of modeling finance may have created a misperception of risk.
Although continuous time finance professors like Merton insist that liquid financial exchanges are examples of arbitrage-free markets, this is in dispute. (Many researchers have shown that elementary trading strategies could have made a consistent profit even in the FX market even fifteen years after their results challenging the arbitrage-free assumptions were first published - I believe examples are given in Hull's Options, Futures, and Other Derivatives, and more advanced textbooks).
Some economic models are written into banking regulations despite the fact that they are purely theoretical. For instance, I personally work on a system which is mandated to project differential exchange rate drift through the UIP model for thirty-year simulation horizons, even though this hypothesis has been thoroughly rejected by all econometric regressions, and no hedge enforcing it is in place.

CSTAR WROTE: (About my mocking of economic models assuming immortality):

Are you making some reference to macroeconomic models such as permanent income hypothesis which try to include life-cycle calculations into the model? This comment is far too vague to be of any use here, and should be in some other article on macroeconmic (sic) models.

My response: Here I disagree with the wording of CSTAR's critique. Although my statement is clearly designed to sound contentious (in the hopes of interesting the reader) it was not, in fact vague - which has a fairly specific meaning, not applying to the claim: "some consumption models treat people as immortal". To be vague it must be unclear whether a claim is true or false. Although the claim "some consumption models treat people as immortal" cannot be falsified it can be proven true (as far as anything else on Wikipedia can be proven). For a book on the subject of immortality and economic models, containing further references, please see: [2], or essays in: Discounting and Intergenerational Equity ISBN 0-915707-89-6

CSTAR has (rightly) subjected my English to forensic analysis, so I feel entitled to question his use of the word "vague" in its colloquial sense. It might be argued that there is a difference between the standards required on the article pages and the talk pages, but according to official Wikipedia guidelines this is false: so long as deliberate mistakes are not made, contributions with typos are still a net benefit, and (therefore) so are edits the using colloquial expressions such as "butterfly effect". The reasoning behind this policy is that anyone can edit the article to make the necessary improvements. However, in the same way that my language may be imprecise despite raising a valid point, his reasoning is valid despite a questionable way of phrasing it: This line about the permanent income hypothesis could profitably be adjusted and moved elsewhere (for now - delete it if unhappy with its presence).

Please feel free to delete or move my changes

Some other contributors to this page seem to have much stronger opinions about it than I do, and since the article seems to be in good hands I am happy to leave it alone. Please don't worry that I'll disrupt any further changes you need to make. From the points above it seems as though CSTAR has attempted to make an edit but found my changes unsalvageable or misplaced. As long as nobody don't wrongly suggests that they are simply opinion (or factually incorrect) I am happy for them to be moved to other articles, or deleted (otherwise I wouldn't have made any submissions).

Please don't worry about deleting my actual text so long as you address the points I have raised in some way (somewhere in Wikipedia, hopefully expressed in a clearer, more precise form).

The rewriting of ideas is exactly what I'm hoping to provoke, and hopefully we can collaborate on other pages in future. My suggestion for a way forward (other than the idea of a generous collaboration) is that new articles should be started, or updated with names relating to a specific aspect of the theory and history of economic modeling (in this way it will be clear what should be included and excluded from each). Some ideas that pop into my head are:

McCloskey critique
Post-autistic economics
Chaos in economic models

Cheers, and if you’ve read all the above, thank you very much, Wragge 00:25, 2005 May 22 (UTC)

Summary of long answer

The very long answer given in the heading above is presumably too long for most people to read. In a way, this sums up one of the main points I was trying to make with it:

An article as general as "Model (economics)" cannot deal at length with any specific point, or spell out what is meant by every off-hand remark. Perhaps the best answer to this is not scientific rigour, but an off-hand reference to another article, simply suggesting how the fields are connected. Criticizing an addition because it lacks rigour might not be the best way of handling a page which covers such a wide range of ideas, because this will lead to either:

The inclusion of too few points (not encyclopedic)
The page becoming massively long.

That is handled by writing in Wikipedia:Summary style. Where a section in the main article, such as this model article, covers that section's topic in the most accurate, readable, and concise way possible, but points with a Main article: foo heading to a longer discussion.

We're not criticizing, we're saying you can do both. If you know this subject, you can add needed material, and have it be backed by other sources. And don't think that asking you to do better is a bad thing. In fact, this article wasn't getting much editing before your additions, so I at least was glad to have them, I'd just like to see them be helpful, not harmful to the overall quality of the article. - Taxman 04:45, May 22, 2005 (UTC)

It is true that in the long term, everything in Wikipedia should be rigorous (having been apportioned the correct amount of space to be written in this way) but, for now, an article like this is a good temporary holding area for aspects of a general topic (this prevents repetition of a point between a main article sub-heading and a stub).

Thanks for reading, Wragge 00:37, 2005 May 22 (UTC)

Every article can approach perfection by keeping both rigor and readability as its primary goals, and continually trying to balance them. No need to entirely give up one to get the other. Please do keep contributing, I'm just suggesting a way for your contributions to be even more successful and impactful. - Taxman 04:45, May 22, 2005 (UTC)

Categories: Talk archives