UnRisk FACTORY 2.1 IS Released

26-Aug-10, UnRisk took UnRisk FACTORY 2.1 to financial institutions.
In May-08 UnRisk FACTORY was released. Version 2.1 is its 6th release since. It supports now all deal types, models and utilities that have beed added to UnRisk 4.1 -recently released.
A fast growing number of banks, capital management firms, insurance, .. enjoy the following benefits from using UnRisk FACTORY: Faster Time-to-Insight - High Accuracy - Complete Evidence - Wide Access - Low Cost of Ownership
more: Release Announcement

Alexander Dumas' Model

In the LikedIn Group Quant Finance we have a thread What do people think of new "Black Swan" funds and "Black Swan" hedging. With others, I share the view that buying insurance against a Black Swan event, illiquidity might become the Black Swan of a Black Swan event (nobody left to pay if it happens).
However, inevitably the discussion went into model-complexity and so on. 
Maurice, a french quant introduced his idea of an Alexander Dumas meta-model. 
He explained: provided I am dueling D'Artagnan (from 3 musketeers). If my sword is too light, I lose and if it is to heavy as well. The same, if I am not familiar with it's usage. 
The only chance I have to beat my opponent is if I am skilled and I carry a sword that is perfect for my skills.


I add: if your sword is perfect in handling, it also needs a reliable and robust material, sharp edges ... (which you understand from its recipe, crystal structure, grinding procedures,... and scenario tests). 
And you need a maker that not only trains you in the usage, but also never gives you false comfort on its material properties, strengths and limits.


And don't apply the motto of the musketeers: all-for-one & one-for-all: one-size-fits-all approaches fail in quant finance.

The Future That Never Happened

Cover story in Wired, Aug-10 .
We still don't have, Laser Guns, Food in a Pill, Robot Servants, Nano Technology, Fusion Power, Quantum Computing ... and Self-Driving Cars.
The reasons are manifold, from technical or economic barriers to the fact that some ideas have never been positioned in any demand landscape (do they relate in any way to how people behave, their needs, wants and frustrations) and if, reframe an opportunity space, looking into it with the eye of potential customers, markets and industries.


Self-Driving Cars. To traffic-weary drivers, it sounded great ... Cars could drive closer together, increasing highway capacity and fuel efficiency while reducing accidents .. Prototypical demand-first innovation?
I also think it is not too hard to make vehicles that automatically adjust their speed and path to the traffic "neighbors". But it might require a lot of pricey  road sensors.
However, risks are another concern - imagine the vehicles, that behave like swarms, "panic" and turn into avalanches ( Starling Flocks .. ).  Imagine the lawsuits after mass-accidents ... So let them be restricted to 50 mph? Sounds not so great then?


This is what comes in my mind, when I think of high-frequency trading running data feeds into black boxes containing some algorithm that makes local decisions ....  that might become horrible in interplay? 
Is it the future that already happened, but should not?

The Future Is Getting Old Like The Rest Of Us?

A film conceived in the format of a TV play, part of The Skills Exchange Project at the Serpentine Gallery in London.

I like this small gallery - see Design Real .

The title came into my mind, when I read about and in The Future Of Finance of the London School of Economics.

I am not an economist and it is hard for me to digest, but reads as fascinating stuff.  I think it is important to understand systemic risk and create theories that help to detect bubble forming. 

And yes the Minsky Hypothesis (the danger of unbalanced hedge-, speculative-, Ponzi-borrowers) is brilliant, and the future isn't like the past.

I confess, I easier read things in the language of mathematics ... like Asset Price Bubbles .

Is it just the understanding of systemic risk and regulatory consequences?

I find we still have home work to do on the microlevel. 

In short, putting numbers on good ideas to make them more precise, testable, actionable, improvable and interpretable, and avoid traps that make them more impressive, powerful and dangerous.

I have often pointed out that each single task in quant finance bears the danger of fundamental mistakes that become horrible in interplay. But there is hope to avoid them.

Let us combine forces to get new fresh ideas ... for a young future avoiding a new Risky Horror Show 

Shanghai Noon

In about a month UnRisk will visit Shanghai to highlighting the latest directions and technologies in Shanghai Seminars 2010 for financial institutions -  together with our partner China Quants .


We want to point ou that there are reasons to introduce mathematical schemes that are not common in financial circles and that it matters to implement them properly and, probably most important,  tie them together to know-how packages.
And we want to point out that each single task can become complex and bear the danger of fundamental mistakes , why those can become  horrible in interplay and that there is hope to avoid them -and how.


Participants will come away with a comprehensive understanding of UnRisk's key capabilities, core design principles, coverage, models and methods. No UnRisk use experience is required.

A typical Seminar Agenda .

1977, The TRS-80 Was Announced on 3-Aug

"Wired", This Day in Tech , Tandy Corp of Texas announced the manufacture of the first mass-product personal computer. The TRS-80 was a desktop machine woefully underpowered by todays standards - 4 KB of RAM, a built-in cassette-based data recorder, a BASIC interpreter ...  and oh, yes it came with Blackjack and Blackgammon.
This was the dawn of the personal computers age ....


Todays standards? Power users are now empowered by multicore machines and CUDA frameworks increasingly boost number-crunching. In Taming The Machine Infernal I wrote about our tests with various solvers in UnRisk.
Unlike others, we do it the UnRisk way avoiding quick-releases and now make comprehensive testing on the next generation CUDA architecture, code named Fermi with improved double precision treatment that is indispensable in calibration  and optimization of financial models and objects. Oh yes, it improves and boosts ....
I keep you informed.
Will history record Fermi as significant milestone? Yes, IMO.

The Botox Economy

Satyajit Das, a long-time author in Wilmott magazine found this brilliant metaphor in his article in the May-10 issue. It is about the flood of money from central banks and governments that has temporarily covered up unresolved and deep-seated problems - financial botox. The surface is glossy and smooth, but the interior ... ? (Botox, commonly used in cosmetic procedures).
For me it is a title-of-the-year candidate. 
At least here in Europe, financial institutions have often used part of the flood and invested in "secure" deals - state bonds, fueling additional state deficit spending that finally led to a redemption "hysteria". 


I do not want to comment on that because so many experts have ... 
But what is interesting to me from the risk modeling point of view: are people aware that credit risk has not only a default, but also an early redemption component? A deeper look into the fiat money system suggests that this has a potential for a systemic risk ... the botox economy risk?


p.s. now on Das's Blog  Botox Economics - 1 and  Botox Economics - 2