Please Models, Just Die

This thought has been lingering with me for awhile, but I am very much hoping that one good thing from the current economic turmoil could be that finance and economics professionals throw most of their models in the trash. I read an interesting NYT times article on how Nouriel Roubini has gone from being an obscure economist to one who now acts as an advisor to governments on the current crisis. Notably, it also highlights that he is not the econometric sort of economist but more the historical perspective and comparative type. You'd thus hope that economic thought, especially in the private sector, might shift away from those absurdly stupid econometric models. I can understand how those models make good talk and are a simple way to cover one's behind. ("my assumptions are x = this, and y = that" looks more rigorous than "well I think this will happen based on anecodotal evidence and history") But let's hope that a lot of people no longer require econometrics talk in order for someone to be deemed a professional economist. Obviously big names have been able to get away with eschewing econometrics, but the every day aspiring economist has had a harder time when fighting for a job on wall street.

The same goes for risk modelling and stock valuation, trading. With all the blow-up and catastrophic losses by mainstream professional thinking, let's hope that models get thrown in the trash here as well. Let's hope that in a job interview if someone says they rarely use models, then it be deemed a strong point rather than a "lack of professionalism". Because the unfortunate problem is that people with models have tended to beat out those who just use good logic and wisdom in the professional sphere. They have looked smarter to those who don't look too hard. Until now. As said above, models make for tight, simple "reasoning". All the dots look connected and there are no apparent loose ends... again, if you don't think too hard. Let's hope that people now get increasingly ridiculed for using models given the absolute failure of some of the best on the street who were drenched in modeling. Obviously some models are good, but I think its understandable that what I mean is, in finance and economics there has been far too much modeling going on and far too little thinking. So please, if anything good can come of the current economic blow- up, then please, models, just die.

Comments

[m3]

here, here.

nassim taleb did an interview recently on bloomberg, where he compared this nonsense to early medicine: the products doctors used often ended up killing the patients.

many lives could have been saved if they just admitted that they didn't understand what they were doing. and in some cases, doing nothing would have been best for everyone.

i wish these econometric/quant guys would realize this. but as you said, their precious time can't be wasted with the wisdom that history teaches us.

at least we can be thankful that these men aren't doctors, because we'd all be dead.

[Alex]

This is a classic case of throwing out the baby with the bathwater.

Almost any model can be great in the right hands, and even the best model will produce garbage if the person using it doesn't know its limitations, assumptions, etc.

[Joseph Weisenthal]

Great post. I agree with the second commenter that there is almost certainly some value in models. But: They really do give the illusion that some of this stuff is hard science.

In real science, you can start off with certain premises, like the speed of light, or the rate of gravity or whatnot and be somewhat certain about what comes out. But as you state, when it comes to economics, the premises are much shakier. And yet it still looks like "science".

Ultimately, this is a good argument for the Austrian School of economics, and the notion that as much as possible, economic ideas should be expressed in verbal logic. And that if you can't express your ideas that way, then you should go back to the drawing board.

[Vincent]

In regards to throwing out the baby with the bath water, I agree that some models are necessary, in fact a model just means basically putting some math down into an excel sheet which surely is always needed to some degree. What I am ranting against is the widespread dependence on and confidence in complex excel models to do analysis and make decisions. I am also ranting against the widespread perception that in many cases using a complex model is more rigorous than a simple model, or just plain logic. I think many of these models actually trick very otherwise smart people into overestimating their accuracy. They start to see everything from the perspective of modeling which means they tend to stop thinking from other perspectives. In many cases I would argue that forecasting models are so full of potential error as to be better off left unused. With these high-error models you can be at best wasting your time and at worst leading yourself down a path of false assumptions which is worse than just not having a model to begin with. While some applications surely exist for models, there are tons of high profile applications out there which are riddled with error and qualify for garbage-in-garbage-out. I would bet that most stocks' financial forecasting models at major investment banks would fit this bill. The funny thing is that while many people will agree that they are garbage, substantial time and effort continues to be devoted to them and the extensive model-users continue to be seen as more intelligent and professional than the model-skeptics by the industry at large.

[nick gogerty]

models are the latin scriptures in the church of finance. Those who don't understand math just nod along, those who do, should be crying heresy.

[Vincent]

Haha, that's a great way of putting it.

[Keith Hazelton]

The last models to die will be FICO and other credit scoring models, the holy grails of the consumer finance industry.

[buy viagra]

Since trash bags are used in American homes as a means of garbage
disposal I have been wondering, how much trash in a landfill is
actually the garbage bags we use to throw the trash out?

[louis vuitton]

They start to see everything from the perspective of modeling which means they tend to stop thinking from other perspectives. In many cases I would argue that forecasting models are so full of potential error as to be better off left unused. With these high-error models you can be at best wasting your time and at worst leading yourself down a path of false assumptions which is worse than just not having a model to begin with.

[generic viagra online]

A very nice blog.I get always something that is new and informative.The author should be appreciated for the knowledge and his research in the field.

[Designer Louis Vuitton handbags]

models are the latin scriptures in the church of finance. Those who don't understand math just nod along, those who do, should be crying heresy.www.baghow.com

[replica watches]

Dear friends,I am a fanatic watch collection, especially the well-known watches, you also can do, just click on my name!!!!!!!!

[oil paintings]

Thanks for a great time visiting your site. It's really a pleasure knowing a site like this packed with great information.

[online savings account]

Financial assets, known as investments, are financially managed with careful attention to financial risk management to control financial risk.

[ugg 5819]

Put your heart and soul into the things what you do that you will feel more easy than before. This is my idea that carry out many years ago.

[small chanel]

Financial assets, known as investments, are financially managed with careful attention to financial risk management to control financial risk.

[coach outlet]

I thought it was enjoyable at best.*

[coach bags]

Yeah, you're right.*

[Jordan Sneakers]

So interesting posts,you wrote the headline you gave,and the meaning was so excited.I am sure people who all like them,and better then,good luck!

[Supra Shoes]

Insist on a campaign day,thanks

[Air Jordan Shoes]

Blessed be God, the Father of Christ Jesus our Lord, who in Christ has blessed us from heaven with every spiritual blessing.

[buy on taobao]

I can imagine TONS of uses for this

[GHD Styler]

*haha I love you post excellent

[air jordan]

That's not his. Steve's is black*_*

[nike air max]

I was once asked.*_*