Serving the Quantitative Finance Community

 
User avatar
Paul
Topic Author
Posts: 11263
Joined: July 20th, 2001, 3:28 pm

The largest act of fraud or vandalism in history?

May 6th, 2020, 11:32 pm

https://lockdownsceptics.org/code-revie ... ons-model/

Passing yourself off as an expert, advising governments, knowing that your advice is flawed,...

Is it fraud, vandalism, what is it? Other than very big!

Surely it’s more than just a cock-up if you are using a predictive model that you aren’t able or willing to fix.
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 10:28 am

First impressions:

I had a quick initial read of this damning report. It does not surprise me one iota. It's what i said all along. It is not for nothing that alarm bells went off when the BBC News showed the ODE models, 17 March 2020. Again, blinding us with science. AMATEURS.
Even my Baruch MFE and MSc quant students (who are very good) in Finance know the issues.

As I said already, you need a latter-day Fred Brooks, not an academic and the university should not never ever be the 'driver' of this project anymore. 

I'll go through the reviews in more detail.


Image
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
bearish
Posts: 6436
Joined: February 3rd, 2011, 2:19 pm

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 1:46 pm

The author makes the recommendation that “all academic epidemiology be defunded”. I don’t really see anything in the supporting argument that is peculiar to epidemiology, so I think the real recommendation (which I know is close to some hearts around here) is that all academic research be defunded. Or, at the very least, all that involves the writing of computer code. Some art history work is probably safe...
 
User avatar
katastrofa
Posts: 10075
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 3:16 pm

Recalling the whole paragraph, I largely agree with it as a mathematical modeller working on similar projects:
"On a personal level, I’d go further and suggest that all academic epidemiology be defunded. This sort of work is best done by the insurance sector. Insurers employ modellers and data scientists, but also employ managers whose job is to decide whether a model is accurate enough for real world usage and professional software engineers to ensure model software is properly tested, understandable and so on. Academic efforts don’t have these people, and the results speak for themselves."

There are people in the scientific community, who realise the problem and work to improve the standards, though.
 
User avatar
Paul
Topic Author
Posts: 11263
Joined: July 20th, 2001, 3:28 pm

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 5:03 pm

The author did rather spoil things with last recommendation. Just makes it look like it's political. 

My small experiences of these government panels is that they get set up with an outcome in mind. We'd be better off with a random selection of experts from many walks of life. Here not just university epidemiologists but a balance of academics, private sector, epidemiologists [I hate typing that word!], mathematicians, economics, actuaries, business people, etc. And you need devil's advocates too. 
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 6:01 pm


There are people in the scientific community, who realise the problem and work to improve the standards, though.
That is necessary, but more is needed.  You are not even wrong. The scientists who care don't have clout.

It's a software management project now plain and simple where other skills take over. Here is a list of the top errors in software projects.

The ODE etc. part is the easiest part.

Ten Common Project Management Mistakes and How to Avoid Them
 
1. Assigning the wrong person to manage the project.
2. Lack of resources and skills.
3. Doing everything yourself.
4. Failure to communicate properly.
5. Poor project initiation.
6. Lack of clear objective.
7. Wrong estimation of time and budget.
8. Failure to manage project scope.
9. Micromanaging projects.
10. Not following a process.
 
I used to be a requirements analyst in the old days and  the biggest problem becomes (after budget allocation and honeymoon)

11. POLITICS in the broadest sense of the word.

The word 'project' is not heard a lot in academia I imagine.

2. Lack of resources and skills.
In the current case putting a scientists with basic C skills ( ) on a project is the worst case scenario. Most scientists realise 1) they are not software developers, 2) they hand over their prototytpe to a s/w team (like what happened, but too late).
Last edited by Cuchulainn on May 7th, 2020, 6:39 pm, edited 1 time in total.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 6:22 pm

The author did rather spoil things with last recommendation. Just makes it look like it's political. 

My small experiences of these government panels is that they get set up with an outcome in mind. We'd be better off with a random selection of experts from many walks of life. Here not just university epidemiologists but a balance of academics, private sector, epidemiologists , mathematicians, economics, actuaries, business people, etc. And you need devil's advocates too. 

It's one thing being able to write it (use Grammarly?) but try spelling it after 93 pints of Peroni.

www.youtube.com/watch?v=88iGoo_WC9o

Practice before the next cocktail party

The American Pronunciation Guide is devoted to descriptive linguistics--i.e., the study of the internal phonological, grammatical, and semantic structures of languages without reference to the history of the language or comparison with other languages.  We believe that the best way to learn how to say "epidemiologist" (or any other word) is to listen to the pronunciation of our peers.  To help foster such descriptive learning, we provide videos combining relevant clips from popular culture.  
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 6:42 pm

Before the GitHub team started working on the code it was a single 15k line C file that had been worked on for a decade, and some of the functions looked like they were machine translated from Fortran. There are some tropes about academic code that have grains of truth, but \

Do they know Fortran arrays are column-major while C arrays are row-major (doesn't help performance!!)? 

“A Big Ball of Mud is a haphazardly structured, sprawling, sloppy, duct-tape-and-baling-wire, spaghetti-code jungle. These systems show unmistakable signs of unregulated growth, and repeated, expedient repair. Information is shared promiscuously among distant elements of the system, often to the point where nearly all the important information becomes global or duplicated. The overall structure of the system may never have been well defined. If it was, it may have eroded beyond recognition. Programmers with a shred of architectural sensibility shun these quagmires. Only those who are unconcerned about architecture, and, perhaps, are comfortable with the inertia of the day-to-day chore of patching the holes in these failing dikes, are content to work on such systems.”
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 6:49 pm

single 15k line C file
Sounds kind of right. One man/woman project.

I bet kats could do it before lunchtime.

If you use standard C++ libraries (ODE, optimisation, RNG) I reckon you  can remove 60% of the code. Just a guesstimate. And maybe lots of copy-and-paste.
Last edited by Cuchulainn on May 7th, 2020, 7:11 pm, edited 1 time in total.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 6:58 pm

The author did rather spoil things with last recommendation. Just makes it look like it's political. 

My small experiences of these government panels is that they get set up with an outcome in mind. We'd be better off with a random selection of experts from many walks of life. Here not just university epidemiologists but a balance of academics, private sector, epidemiologists , mathematicians, economics, actuaries, business people, etc. And you need devil's advocates too. 
And most importantly, software professionals! Once the project starts most stakeholders are needed as second opinion keep them from pestering the software team every day with new requirements. The team should have a strong leader to protect them from extraneous bullshit.
A big (and not uncommon) risk factor is when the spon$or annoints a (software) project manager to lead the software project. In many cases this manager's background is economics, law or even journalism. Then you know it will be a Death March.
Projects can fail even before a single line of code has been written. Lots of them negative waves.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
katastrofa
Posts: 10075
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 8:08 pm

Not the largest I've seen but certainly enough for two.
Image
 
User avatar
Paul
Topic Author
Posts: 11263
Joined: July 20th, 2001, 3:28 pm

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 8:09 pm

I did LOL, but it’s cruel.
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 7th, 2020, 9:15 pm

I am having problem with what you have written; the superlative form of an adjective is referring to an otherwise unmentioned noun. Or did you mean the comparative form? What went before?
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Paul
Topic Author
Posts: 11263
Joined: July 20th, 2001, 3:28 pm

Re: The largest act of fraud or vandalism in history?

May 8th, 2020, 1:23 am

Not the largest I've seen but certainly enough for two.
Image


Now I've got "Weebles wobble but they don't fall down" going through my head:

Image
 
User avatar
Cuchulainn
Posts: 64417
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: The largest act of fraud or vandalism in history?

May 8th, 2020, 9:02 am

I watched Newsnight last night. They are starting to question the IC models. The models don't work primarily because the government (like USA) reacted too late, did no testing nor contact tracing (they stopped mid-March). In short, there is insufficient data for the model to work properly.

Boris' bluffing didn't pay off.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl