This of course is very much alive in the financial areas and one of the reasons myself and others who know technology think it is absolutely insane that we are looking at anther attorney to run the SEC. Nothing personal but they can’t move fast enough to keep up. This is sad that so many out there like me keep repeating it over and over and for some reason the government can’t see this, sad. We all also have models and code in healthcare. If you look at job openings you will see that Untied for one has a never ending stream of openings for Quants, those who write the models and math which designs the analytics they create and use. The Quants are the next level up from a programmer, as I used to be and the programming portion takes the model and builds the software to execute models and that’s how the algorithms work.
Granted I know this is a little deep here but it is what it is. Stop and think why do we have folks like me warning about this, folks who write models warning about this? It’s real and I have pointed out a few examples and am living one right now with Medicare stipulations in getting a member of my family taken care of. We have too many folks out there that believe in Algorithm Fairies and while analytics helps up make better decisions when done right it also is an easy to make money as most of the world doesn’t understand how this works. It worked well during sub prime days as you saw who make money but there’s also the downside when reality comes crashing in as the world of math and coding does have a side of fiction that can be used to snow all. If you scroll down and watch the videos at the bottom of this blog, or take time and watch the same and additional videos at my Algo Duping page, you will understand a lot more.
We have clashing worlds in healthcare with for profit insurers versus the clinical data which is credible working with and against each other. Fraud can enter at any time when there are big profits to be made. Cathy O’Neil who is in one of the PBS videos below writes about this on her blog. She worked it on Wall Street and saw what happened and what continues to happen if ethics are not in place. Model builders can go wild and create fiction that reduces risk and it happens all the time, and thus so, models need to be verified. This is the problem with the Department of Justice in not having investigators to go this route and investigate and verify validity of code and models. If you happened to watch the PBS special called “The Untouchables, Too Big to Jail” from what is shown here not only was the obvious not touched by there was no mention anywhere of the DOJ even looking at the “models” used to create the sub prime loans. It’s not the guy who wrote the software that all the banks purchased, it’s what they did with it.
“The Untouchables-Too Big to Jail” Frontline Documentary Shows Department of Justice’s Fear of Prosecuting Big Banks–No Confidence In Using Current Day Technologies To Investigate –Video
In her latest post she references yet one more Code of Conduct document, again created by those who want clean and non fraudulent code. Scroll down again to the bottom and watch the documentary “Quants, the Alchemists of Wall Street” and you will see Emanuel Derman and Paul Wilmott discuss their code of ethics they came up with a long time ago. They wrote their Financial Modeler’s Manifesto after the financial crisis…see both tin the Quants video at the bottom of this page. They both tell you how common sense drops and that a major rethink is needed before we have a mathematical market meltdown. If the market melts down with math, so does healthcare as so many are traded on the open markets, same thing.
In her blog writings she found yet one more document from a company who employs Quants and again their efforts to again talk about not cheating and what to do when working for a client who wants one to “work the numbers” for the models. Again this is all over and in healthcare too, and Quants get paid a lot of money for writing models and when things come crashing down, who do they go to, the Quants. Now there’s a point in time too where the Quants say “no” and that’s what this is all about, refusing to write models and code for money with no concern about ethics and the impact on the general consumer. Here’s the document she references and the site looking for comments on their draft. You should read it and below are a few paragraphs that outline what the data scientist should do and respond if the client is pursuing areas beyond what the data substantiates for profit for example.
“Rule 8 - Data Science Evidence, Quality of Data and Quality of Evidence
(a) A data scientist shall inform the client of all data science results and material facts known to the data scientist that will enable the client to make informed decisions, whether or not the data science evidence are adverse.
(b) A data scientist shall rate the quality of data and disclose such rating to client to enable client to make informed decisions. The data scientist understands that bad or uncertain data quality may compromise data science professional practice and may communicate a false reality or promote an illusion of understanding. The data scientist shall take reasonable measures to protect the client from relying and making decisions based on bad or uncertain data quality.
(c ) A data scientist shall rate the quality of evidence and disclose such rating to client to enable client to make informed decisions. The data scientist understands that evidence may be weak or strong or uncertain and shall take reasonable measures to protect the client from relying and making decisions based on weak or uncertain evidence.
(d) If a data scientist reasonably believes a client is misusing data science to communicate a false reality or promote an illusion of understanding, the data scientist shall take reasonable remedial measures, including disclosure to the client, and including, if necessary, disclosure to the proper authorities. The data scientist shall take reasonable measures to persuade the client to use data science appropriately.”
And a few more words here….
“(g) A data scientist shall use reasonable diligence when designing, creating and implementing algorithms to avoid harm. The data scientist shall disclose to the client any real, perceived or hidden risks from using the algorithm. After full disclosure, the client is responsible for making the decision to use or not use the algorithm. If a data scientist reasonably believes an algorithm will cause harm, the data scientist shall take reasonable remedial measures, including disclosure to the client, and including, if necessary, disclosure to the proper authorities. The data scientist shall take reasonable measures to persuade the client to use the algorithm appropriately.”
As you can see the word “algorithms” is spelled out plain and clear. If you visit this blog often enough then you have read about either Algo Duping or the Attack of the Killer Algorithms, the second of which is usually the result of the first. So why does a profession of Quants feel this is necessary? They have worked hard to excel in their professions but yet they are also wise enough to be aware of what greed can do and how they can be coerced. We have a long history of it on Wall Street and again watch the video about the Quants of Wall Street. The folks speaking out here came from the old school when they were hired to create efficiencies for banks and companies and we are way beyond efficiencies today as models make money when combined with algorithmic code that executes and not much changes without model and code updates.
Sure they get paid well but nothing compared to what the front offices make and they are at times end up being what some call the “grunts of code and models” as none of this happens without their talent. They don’t want to be jailed or brought to trial over models and math that is written that is unethical either, they want to create models that are as accurate as possible and bring clarity and good decision making to light. You can read their comments at the document referenced above and at other places on the web. They see the storm brewing and you can’t blame them for speaking out and not wanting to be put in a situation with using unethical models and code. If you have knowledge in the quantitative area, they are looking for comments for others to add as they call this a “draft” as it stands now. Here’s another good post to read about the complexities of data today and the call for bankers, accountants, modelers and others to be one on one with honesty.
I felt it necessary to bring this topic up again as everything operates off business models and algorithms out there today. Laws need to have these types of rules lined out today and we can’t rely on tons of text to do it, it’s not that kind of a world anymore. One of our own weaknesses is the fear of math and that is so profound and sure are times I don’t want to do or think about math either, but there are those that use for profit so our fears and lack of knowledge from the top all the way down allows dirty code and bad models to exist without anyone ever asking a question. DOJ needs to bone up on this as that’s where the heart of their investigations lie as nothing changes until the execute code is changed on the servers running 24/7.
Algo Duping” – PLOS One Journal Publication Explains Why The Fear of Math Plays a Big Role As One Underlying Reason We All Get Duped And Those Who Don’t Fear Math Take All the Money, Gradually, Using “Mathematical Formulas & Algorithms”
Models can be used out of context and marketed that are marginal or knowingly not verified to be accurate enough to produce substantial and accurate results. This brings me around to all the data selling that is going on out there. It’s a virus and banks and companies are making billions in profit. Once data folks get their hands on data bases they go to work with either linear or non linear methodologies and create analytics. It’s the way it works and has for years. This presents yet another danger of flawed data and we are all seeing it grow out there. Models and algorithms can be written to find fraud and they can also be written to create fraud.
Bill Gates has come out and said the same thing when it comes to healthcare about the “models” used with stating that more money is spent on curing baldness than tackling disease. So here we go again, it’s all around us if we want to pay attention and listen as markets are all algorithms and models.
“Our priorities are tilted by marketplace imperatives,” he said at the Royal Academy of Engineering’s Global Grand Challenges Summit. Mr. Gates, who is one of the world’s richest men with an estimated $67bn fortune, said governments should act to offset what he described as this “flaw in the pure capitalistic approach”. His comments will be interpreted as another blast at the large pharmaceutical companies, which have long been criticized for ploughing money into developing “lifestyle drugs” and neglecting research that could save the lives of the world’s poorest.”
We have data scientists trying to figure out what to do with the data and in some cases it’s pretty easy, like registry information in healthcare, and that’s a good thing for sure as shown by Kaiser Permanente. As a matter of fact some of the models written in other areas can benefit healthcare too, valuable analytics. So again when we see the activity of the Quants themselves with a “code of ethics’ should you not stop and think?
I still go back and revisit the Russian coder situation with Goldman and how that was never done right and we have people who know nothing about this complexity trying to hold court! I sad you better sent the jury to Code Academy so he could get a fair trial. No way can the person off the street understand all the complexities of what Goldman accused him of. I’m not saying innocent or guilty but tech folks like want “the code truth” and not one more fantasy out there. You know what if I were a Quant I would be working on a code of ethics too because they know that temptations and offers will be there from those who are less than honest and go about their business of hiding risk to profit. It’s not like this is a big secret either it’s just that everyone is bliss.
Goldman Sachs Programmer Who Went to Jail for Stealing Code Has His Conviction Overturned–You Can’t Get A Jury of Peers Off the Street for Crimes With High Tech Algos
Now to move on this is another post yet talking about the value and validity of models and algorithms…used to make analytics. A bunch of bankers from Australia made this observation originally and I agree with them. The Algo Fairies that HHS, CMS, the Heritage Foundation (just to name a few) and others think that fly around out there to give black and white answers belong at Disneyland. You are going to get marketed in healthcare and everywhere else and the big job is going to be to find value and accuracy. You’ll get sucked in as we all do at times.
Half of Analytics Investments By Companies and Banks Will Be a Waste–What Do We Analyze with Big Data and Does It Have Value–Some Algo Fairies Would Do Better at Disneyland…
If you want to see a great video with NASA and several other large entities stumbling around with value with big data, watch the video below and here’s a forum and yes they are talking right up front about their Quants they have working for them. Quants work all over and not just at banks. You can read my entire post here but here’s the video below, watch it as flawed data grows and data content can be used “out of context”. You won’t know as you have been marketed so well and still think the US has enough money to rewrite Medicare with vouchers:) Goodness I wish we had Congressmen that understood the cost of IT infrastructure as the cost to do such would probably run 2nd to the budget of DOD by the time it was done in about 8 years or so:)
So you want to stop inequality…get to the root and stop the outrageous data selling that goes on that feeds the greedy process above with analytics created to make money without any concern on how it hits the consumer, that’s where the control is, not politicians and I wish they would get educated here, all of them before they send us further down the river. Unless this is somehow regulated with a DOJ that has no clue on how to investigate models and high tech crime, we are screwed.
Time Has Come to License and Tax the Data Sellers of the Web, Companies, Banks, Social Networks..Any One Making a Profit-Latest Microsoft/Google Privacy War Helping the Cause –Consumers Deserve to Know What Is Being Sold and To Who in a Searchable Format
Now after all of this do you understand why the Quants want a code of ethics? They are not the bad guys but they have the talent and it’s the code and models they write that controls all. Over three years ago I made this post…do we need a department of algorithms? I saw it then so did many others but nobody wants the education process out there as it messes up the game of code for cash.
Does the US Government Need a Department of Modeling and Algorithms–Is Data Addiction and Abuse the Next Up and Coming 12 Step Program, Some Classic Posts & Topics Revisited
Do you like the fact that insurance policies keep going up? Well the money has to come from somewhere to keep their quants and algorithms developing and they are a mixed bag at times as they do produce value in some areas and mess up in others, like the United Healthcare algorithms that short paid doctors and hospitals for 15 years…algos for money until someone questioned the code.
Insurance Companies Are Buying Up Consumer Spending Data-Time is Here to License and Tax the Data Sellers-As Insurers Sell Tons of Data, Gets Flawed Data When Data Buyers Uses Out of Context Too
Again in the Quant video you can hear Paul Wilmott stating “Quant fix the numbers, get the risk down”…would you love to be a fly on the wall over the last number of years to hear how many times that request was probably made? When there’s billions to be made as they did with sub primes, I bet it was a lot.
So to sum this up when you think about and read about Big Data, give this some thought on how the queries, data and algorithms are being created. If something sounds fishy, then question it. Watch the first video below from Charlie Siefe at NYU and he will cue you in with a lot of knowledge on the Algo Duping side of this and how you get marketed and fooled. As he says, “numbers don’t lie but people do”. Again I thought this was important to talk about Big Data and let all know that all analytics are not accurate and not useful and some can fool you. There’s no Algorithm Fairies to give you black and white decision making power all the time today but there is good data to help us make better decisions. So when you hear Big Data next time, think about it as it is not always the best thing since sliced bread and there’s a long ways to go, as again, data and models can catch fraud and they can create it.
Consumer Mobile Health Monitoring Has Issues That Stifle Faster Growth, Epidemic Data Selling In Healthcare and Unrealistic Immediate Expectations With Patients Changing Their Behaviors
So take this topic of a Data Science Code of Ethics seriously because they wouldn’t be talking about it if the Quants themselves were addressing within their own field. They see it and know it is there with the temptation of big money with models and codes and nobody checks them as we don’t seem to have the talent and I guess enough tech knowledge in government to actively pursue fraud and bad code when it is needed. For some reason or another folks just seem to think that all code is good and accurate..wake up as the experts smarter than me are telling you otherwise, folks with hands on experience and we can certainly stand to get all the transparency we can get our hands on today. BD