recentpopularlog in

jerryking : modelling   9

Coronavirus, Ray Dalio and forecasting in an age of uncertainty
March 18, 2020 | Financial Times | by Gillian Tett.

** Against the Gods: the Remarkable Story of Risk by Peter Bernstein.
** Uncharted by Margaret Heffernan.
** Radical Uncertainty by Mervyn King and John Kay.

Ray Dalio, founder Bridgewater Associates, admitted that he had been caught flat-footed by the recent coronavirus-driven market swings. .... It seems that the systems that Bridgewater developed to analyse the flows of finance and economic activities — which have traditionally driven its bets on the direction of stocks, bonds and other securities — did not offer any guidance when looking at a rare event such as the current pandemic. “We did not know how to navigate the virus and chose not to because we didn’t think we had an edge in trading it,” Dalio went on to explain. “So, we stayed in our positions and, in retrospect, we should have cut all risk.”
Now, many readers may feel baffled by this, given that the whole point of investing with a hedge fund is that they are supposed to beat the markets at times of stress....scorn is the wrong response here.....What is interesting to ponder is what this episode reveals about the nature of forecasting — and our modern attitudes towards time.
......the way we think about time is a defining feature of the post-enlightenment world. During much of human history, the future was viewed as a vague and terrifyingly unknowable blur marked by constant bargaining with deities (to ward off disaster) or cyclical seasonal rhythms (of the sort that underscore Buddhist cognitive maps). In modern, post-enlightenment western cultures, however, a linear vision of time emerged that presumes the past can be extrapolated into the future, with a sense of progression, not just cyclicality.

In the 20th century, this gave birth to the risk management and finance professions, as Peter Bernstein wrote two decades ago in his brilliant book Against the Gods: the Remarkable Story of Risk.
....... By 2000, innovations such as computing and the internet were turbocharging the forecasting business to an extraordinary degree, “Human discomfort with uncertainty . . . has fuelled an industry that enriches itself by terrorising us with uncertainty and taunting us with certainty,” However, while the forecasting business has made its “experts” very rich, it is also based on a fallacy: the idea that the future can be neatly extrapolated from the past. Moreover, the apparent success of some pundits in predicting events (such as the 2008 crash) makes them so overconfident that they get locked into particularly rigid models. “The harder economists try to identify sure-fire methods of predicting markets, the more such insight eludes them,”

Is there a solution? Heffernan’s answer is to embrace uncertainty, build resilience, use “narrative” (or qualitative) analyses instead of rigid models and to respect the wisdom of diverse views to avoid tunnel vision.

..........accept radical uncertainty and rethink our models......models (whether they emerge from computer science or economics) are like a compass in a dark wood at night. Navigation tools can give you a sense of direction and orientation; it would be ridiculous to toss them out entirely.

However, if you rely exclusively on them, accidents occur. If you walk through a wood just looking down at the dial of a compass, you will bang into a tree or worse. The trick, then, is to use navigation aids but also to maintain your peripheral vision..........the insights of cultural anthropology is one way to maintain peripheral vision, since it provides a social context for looking at our favoured tools (and thus a way to see their shortcomings).........Either way, now more than ever, we need broader perspectives — and humility — when trying to assess what might happen next, not just with the markets but with the coronavirus outbreak too.
books  Bridgewater  COVID-19  cultural_anthropology  extrapolations  fallacies_follies  forecasting  Gilliam_Tett  hedge_funds  heterogeneity  humility  linearity  mistakes  modelling  pandemics  peripheral_vision  Peter_Bernstein  predictive_modeling  qualitative  Ray_Dalio  risk-management  resilience  shortcomings  turbulence  uncertainty 
21 days ago by jerryking
How to Prepare for the Next Recession: Automate the Rescue Plan
San Diego 4h ago
As someone with an engineering background (both education and mindset) this kind of simplistic design of complex systems is very concerning.

If anyone remembers Nassim Tal...
complexity  economic_downturn  ecosystems  financial_crises  howto  letters_to_the_editor  modelling  models  Nassim_Taleb  oversimplification  preparation  recessions 
may 2019 by jerryking
The Evolving Automotive Ecosystem - The CIO Report - WSJ

An issue in many other industries. Will the legacy industry leaders be able to embrace the new digital technologies, processes and culture, or will they inevitably fall behind their faster moving, more culturally adept digital-native competitors? [the great game]

(1) Find new partners and dance: “The structure of the automotive industry will likely change rapidly. Designing and producing new vehicles have become far too complex and expensive for any likely one company to manage all on its own.
(2) Become data masters: “Know your customers better than they know themselves. Use that data to curate every aspect of the customer experience from when they first learn about the car to the dealership experience and throughout the customer life cycle. Having data scientists on staff will likely be the rule, not the exception.
(3) Update your economic models: “Predicting demand was hard enough in the old days, when you did a major new product launch approximately every five years. Now, with the intensity of competition, the rapid cadence of new launches, and the mashup of consumer and automotive technology, you may need new economic models for predicting demand, capital expenditures, and vehicle profitability.
(4)Tame complexity: “It’s all about the center stack, the seamless connectivity with nomadic devices, the elegance of the Human Machine Interface.
(5) Create adaptable organizations: “It will take a combination of new hard and soft skills to build the cars and the companies of the future. For many older, established companies, that means culture change, bringing in new talent, and rethinking every aspect of process and people management.
Apple  automotive_industry  autonomous_vehicles  ecosystems  Google  know_your_customer  adaptability  CIOs  layer_mastery  competitive_landscape  competitive_strategy  connected_devices  telematics  data  data_driven  data_scientists  customer_experience  curation  structural_change  accelerated_lifecycles  UX  complexity  legacy_players  business_development  modelling  Irving_Wladawsky-Berger  SMAC_stack  cultural_change  digitalization  connected_cars  the_great_game 
april 2015 by jerryking
You can’t predict a black swan - The Globe and Mail
The Globe and Mail
Published Thursday, Jan. 29 2015

The New York snowstorm that wasn’t, like the Swiss currency storm that was, are reminders that sophisticated computer models used to predict the future are useless in the face of the unpredictable. Instead of seeking a false assurance in the models, it’s better to prepare, to the extent possible, to weather any storm Mother Nature or man dishes up.

Black swans are “large-scale, unpredictable and irregular events of massive consequence,” as defined by the author who popularized the term in a 2007 book. Given their unpredictability, says Nassim Nicholas Taleb, the solution cannot lie in developing better predictive methods....Robust policy – such as sustainable public finances or effective bank regulations – must be designed to withstand black swans.
Konrad_Yakabuski  forecasting  weather  public_policy  reminders  modelling  unpredictability  assumptions  antifragility  Nassim_Taleb  black_swan  resilience  risk-management  policymaking 
january 2015 by jerryking
Meet the SEC’s Brainy New Crime Fighters - WSJ
Updated Dec. 14, 2014

The SEC is mustering its mathematical firepower in its Center for Risk and Quantitative Analytics, which was created last year soon after Mary Jo White took charge of the agency to help it get better at catching Wall Street misconduct. The enforcement unit, led by 14-year SEC veteran Lori Walsh, is housed deep within the warrens of the SEC’s Washington headquarters, and staffed by about 10 employees trained in fields such as mathematical finance, economics, accounting and computer programming.

Ms. Walsh says access to new sources of data and new ways of processing the data have been key to finding evidence of wrongdoing. “When you look at data in different ways, you see new things,” she said in an interview
alternative_data  analysis  analytics  arms_race  data  data_driven  enforcement  fresh_eyes  hiring  information_sources  mathematics  misconduct  models  modelling  patterns  perspectives  quantitative  quants  SEC  stockmarkets  Wall_Street 
december 2014 by jerryking
Big Data should inspire humility, not hype
Mar. 04 2013| The Globe and Mail |Konrad Yakabuski.

" mathematical models have their limits.

The Great Recession should have made that clear. The forecasters and risk managers who relied on supposedly foolproof algorithms all failed to see the crash coming. The historical economic data they fed into their computers did not go back far enough. Their models were not built to account for rare events. Yet, policy makers bought their rosy forecasts hook, line and sinker.

You might think that Nate Silver, the whiz-kid statistician who correctly predicted the winner of the 2012 U.S. presidential election in all 50 states, would be Big Data’s biggest apologist. Instead, he warns against putting our faith in the predictive power of machines.

“Our predictions may be more prone to failure in the era of Big Data,” The New York Times blogger writes in his recent book, The Signal and the Noise. “As there is an exponential increase in the amount of available information, there is likewise an exponential increase in the number of hypotheses to investigate … [But] most of the data is just noise, as most of the universe is filled with empty space.”

Perhaps the biggest risk we run in the era of Big Data is confusing correlation with causation – or rather, being duped by so-called “data scientists” who tell us one thing leads to another. The old admonition about “lies, damn lies and statistics” is more appropriate than ever."
massive_data_sets  data_driven  McKinsey  skepticism  contrarians  data_scientists  Konrad_Yakabuski  modelling  Nate_Silver  humility  risks  books  correlations  causality  algorithms  infoliteracy  noise  signals  hype 
march 2013 by jerryking Living in the real world of finance
December 9, 2011 | G&M | by David Parkinson.
Both a scientist and financial guru, Emanuel Derman warns of relying on mathematical models to predict stock movements. As David Parkinson reports, investors should beware the wild card of human nature...Mr. Derman was in Toronto discussing his new book, Models. Behaving. Badly: Why Confusing Illusion With Reality Can Lead to Disaster, on Wall Street and in Life.

boundary_conditions  finance  quantitative  Wall_Street  Colleges_&_Universities  books  physics  models  mathematics  stockmarkets  biases  modelling  dangers  false_confidence  human_factor  stock_picking  illusions  oversimplification  in_the_real_world 
january 2012 by jerryking

Copy this bookmark:

to read