A significant part of my work is probability modeling, but I have been doing it too long to believe that the model will predict any sort of reality. Why do I continue doing it? Because it seems to make clients happy and they pay me a lot of money to do it — even after I explain that it’s like the end results will look nothing like the model.
The engineer within me asks: "What is the probability.. probability will be useful in practice?"
And the answer is: "It can be useful in keeping accountants in check when you wish to overbuild against an eventuality and they oppose it; Or when you simply wish to build well against all odds... "
More evidence that man's favorite pastime is trying to predict the future through all means; I think God must find it enormously entertaining. Probability is an example of extreme mental vanity and pride.
Over six decades ago, during my first introduction to probability theory, I asked a high school teacher, "What about all of the assumptions you had to make before you started?"
He answered with his usual - "not relevant to this exercise".
Apparently not much has changed since the early sixties.
Nothing has a probability, for those who say probability is uncertainty.
So, I don't have a probabilty of having cancer.
I used to be worried because I took SSRI's drugs for many years. As most people know or have been told, these drugs are supposed to increase the quantity of one chemical between the synapses, and there are many synapses everywhere. The drug also affects the intestines. I thought, "maybe I develop a cancer in the brain near the skull or in the brain near the abdomen because of those stupid poisons I took." That worried me a little, but I'm not superstitious: it brings bad luck.
I am very cynical about politics. I think (too reductionistically, I'm afraid) that politics is all about price control to optimize profits for each oligarchy. Pharma (big and small) is one huge special interest of politics, as everyone knows.
So, the cynicism industry has interest in convincing everyone that frequentist probability is all there is. This determines my life as a consumer-slave. I will get cancer, and I must be scared, otherwise profits will dwindle. Or that is what they want me to live by.
The best way to predict something is to cause it yourself, I've been told.
I wonder if by the simple task of cultivating skepticism I will get better odds at not living in panic and not developing cancer.
What are the odds of slashing cancer diagnoses rates by 95% if half the population stopped watching the News and started learning about statistics and probability?
OK I have read it. Makes sense to me. Maybe I should read your book Uncertainty. I have read and re-read the book Against the Gods a few times, I really like it. I have been getting more into probability as I get older, with some recent practical work I have done in pseudo random number generators and Poisson distributions for measuring radioactive decay. Years ago I seriously resented being forced to take part in brainless Six Sigma training at my then employer, although in hindsight a learned a few things despite it all. I have spent a lot time in my career in engineering R&D managing and mitigating technical risks via waterfall charts and risk matrices - development testing. As I now work in energy policy, I increasingly try to emphasize to my colleagues that when proposing drastic changes to energy systems, the question isn't whether it will or wont work, but what is the probability it will work, and work reliability. I increasingly view the world overall as a sea of uncertainty and probability in which I have to navigate and counsel others how to navigate. So I like reading your stuff.
One further thought. In the engineering world, probability, risk and uncertainty is the water in which we swim. Most of the risk we deal with is unquantified and mostly unquantifiable. There are some niche areas in machine reliability where Weibull analysis is used, and other reliability metrics, but most risk is simply semi-quantified, in likelihood + consequence matrices. The idea that probability/uncertainty can't be valid if it isn't quantified is strange to me.
The penny began to drop for me years ago when I tried to grasp the mystery of “probability”. I realized that every event was unique and its probability, in retrospect was always 1 (it’s not an event if it didn’t happen).
You can aggregate events by losing information — the information that distinguishes individual events. So you make real events into abstract, unreal, events and sometimes you can make statements about these abstract, unreal events that can help you formulate predictions about the concrete events your abstraction (model), in your mind, represents, provided your abstraction didn’t lose too much information.
A supervisor once asked me what're the chances of that software bug happening in the field? After I gave some mumbo jumbo answer, he answered "100 percent: it's going to happen to someone, somewhere, sometime." And with that, I give you:
"the reason we cannot use relative frequency is there is no fixed information or set that can be used to make a ratio for either proposition. Of what divided by what should the probabilities be calculated? "
The whole post is fantastic but I want to draw special attention to these lines.
In a philosophy of science seminar years ago, some eager grad student was going on about the fantastically low probability of our universe existing at all, with the apparently fine-tuned laws that it has, compared to "all the other possibilities".
Being confused, since I only know of the one universe that we actually live in, I asked him something like, "probability compared to what?" All the "possible worlds" nonsense is sketchy modal logic plus even sketchier physics. He didn't like my question.
The image at the beginning of the post is fascinating, because it's complete gibberish. Obviously AI generated, the "intelligence" clearly does not understand any actual mathematics. Instead, it views all mathematical notation as strange glyphs, and so it generates its own strange glyphs to mimic the real thing. It would interesting to see how many people actually notice this.
<--nit picker (and one who also used to be a big fan of p-values)
Briggs, I like to nag you because I admire your love of ideas and your high-level of critical thinking. That being said, "what say you" to this proposition:
"There is an existing probability of distilled water boiling when brought to 100 degrees Celsius while at the atmospheric pressure associated with being at sea level. There is a second probability -- also one which "exists" -- of that same water boiling at 50 degrees Celsius at that same atmospheric pressure."
Did I just lie, or did I just tell the truth?
Does water really have this "natural propensity to boil under heat" that I claim that it does?
I’d be interested in the online class… partly because I almost failed a class on probability at a service academy 50 years ago. Had not the prof, due to his evident frustration of my density, basically given me the problems on the final during one of our many extra tutoring sessions I would have failed.
I believe the “credentialed class” use “studies” based on probability to manipulate the masses, just like the FDA and CDC use “studies” to manipulate the masses.
A significant part of my work is probability modeling, but I have been doing it too long to believe that the model will predict any sort of reality. Why do I continue doing it? Because it seems to make clients happy and they pay me a lot of money to do it — even after I explain that it’s like the end results will look nothing like the model.
The engineer within me asks: "What is the probability.. probability will be useful in practice?"
And the answer is: "It can be useful in keeping accountants in check when you wish to overbuild against an eventuality and they oppose it; Or when you simply wish to build well against all odds... "
Ah, it's ugly, but I like it.
More evidence that man's favorite pastime is trying to predict the future through all means; I think God must find it enormously entertaining. Probability is an example of extreme mental vanity and pride.
I wonder what may be the PROBABILITY that the "vaccine" currently being pushed by various "authorities" is actually "SAFE & EFFECTIVE"
.
Question EVERYTHING
.
.
.
.
WHY?
Over six decades ago, during my first introduction to probability theory, I asked a high school teacher, "What about all of the assumptions you had to make before you started?"
He answered with his usual - "not relevant to this exercise".
Apparently not much has changed since the early sixties.
Nothing has a probability, for those who say probability is uncertainty.
So, I don't have a probabilty of having cancer.
I used to be worried because I took SSRI's drugs for many years. As most people know or have been told, these drugs are supposed to increase the quantity of one chemical between the synapses, and there are many synapses everywhere. The drug also affects the intestines. I thought, "maybe I develop a cancer in the brain near the skull or in the brain near the abdomen because of those stupid poisons I took." That worried me a little, but I'm not superstitious: it brings bad luck.
I am very cynical about politics. I think (too reductionistically, I'm afraid) that politics is all about price control to optimize profits for each oligarchy. Pharma (big and small) is one huge special interest of politics, as everyone knows.
So, the cynicism industry has interest in convincing everyone that frequentist probability is all there is. This determines my life as a consumer-slave. I will get cancer, and I must be scared, otherwise profits will dwindle. Or that is what they want me to live by.
The best way to predict something is to cause it yourself, I've been told.
I wonder if by the simple task of cultivating skepticism I will get better odds at not living in panic and not developing cancer.
What are the odds of slashing cancer diagnoses rates by 95% if half the population stopped watching the News and started learning about statistics and probability?
OK I have read it. Makes sense to me. Maybe I should read your book Uncertainty. I have read and re-read the book Against the Gods a few times, I really like it. I have been getting more into probability as I get older, with some recent practical work I have done in pseudo random number generators and Poisson distributions for measuring radioactive decay. Years ago I seriously resented being forced to take part in brainless Six Sigma training at my then employer, although in hindsight a learned a few things despite it all. I have spent a lot time in my career in engineering R&D managing and mitigating technical risks via waterfall charts and risk matrices - development testing. As I now work in energy policy, I increasingly try to emphasize to my colleagues that when proposing drastic changes to energy systems, the question isn't whether it will or wont work, but what is the probability it will work, and work reliability. I increasingly view the world overall as a sea of uncertainty and probability in which I have to navigate and counsel others how to navigate. So I like reading your stuff.
Many thanks.
One further thought. In the engineering world, probability, risk and uncertainty is the water in which we swim. Most of the risk we deal with is unquantified and mostly unquantifiable. There are some niche areas in machine reliability where Weibull analysis is used, and other reliability metrics, but most risk is simply semi-quantified, in likelihood + consequence matrices. The idea that probability/uncertainty can't be valid if it isn't quantified is strange to me.
I am going to study this in detail when I have time this week.
Try not to eat before doing so.
May be your best one liner. Ever.
Thanks for the laugh.
You kill me.
P(time this week)?
This should be fun, Matt. Looking forward to it.
I'm going to try to keep up while I litigate.
Very nice work!
The penny began to drop for me years ago when I tried to grasp the mystery of “probability”. I realized that every event was unique and its probability, in retrospect was always 1 (it’s not an event if it didn’t happen).
You can aggregate events by losing information — the information that distinguishes individual events. So you make real events into abstract, unreal, events and sometimes you can make statements about these abstract, unreal events that can help you formulate predictions about the concrete events your abstraction (model), in your mind, represents, provided your abstraction didn’t lose too much information.
Anyway, that’s how this layman interprets it.
A supervisor once asked me what're the chances of that software bug happening in the field? After I gave some mumbo jumbo answer, he answered "100 percent: it's going to happen to someone, somewhere, sometime." And with that, I give you:
https://www.youtube.com/watch?v=sD5As76BEP8&pp=ygUXY29tbWFuZG8gc29uZyBzb21ld2hlcmU%3D
https://ito33.fr/articles/1011_ayache_0.pdf
Thanks.
A good rule of thumb.
Wonderful William! Loved every word of it. Yes, please do a class.
Say, what kind of math was that kid looking at? Is he an alien?
Of course.
"the reason we cannot use relative frequency is there is no fixed information or set that can be used to make a ratio for either proposition. Of what divided by what should the probabilities be calculated? "
The whole post is fantastic but I want to draw special attention to these lines.
In a philosophy of science seminar years ago, some eager grad student was going on about the fantastically low probability of our universe existing at all, with the apparently fine-tuned laws that it has, compared to "all the other possibilities".
Being confused, since I only know of the one universe that we actually live in, I asked him something like, "probability compared to what?" All the "possible worlds" nonsense is sketchy modal logic plus even sketchier physics. He didn't like my question.
Mostly no, probability exists as a thought. As a quantifiable measure of all your information.
The coin flip video is great, and shows a coin flipper with some high probability of flipping a coin to be heads. But is it 80%, 90%, 95,99,99.9%?
Would you really bet on a head at 9:1 odds, or 99:1 odds for $1000 (99,000 if not H).
All decisions are made with uncertainty, described as probabilities, analogously as frequency measures but different.
(Cc comment from coin flip machine.)
The engineering dictum 'all models are wrong, but some are useful in certain circumstances' applies to probability.
The image at the beginning of the post is fascinating, because it's complete gibberish. Obviously AI generated, the "intelligence" clearly does not understand any actual mathematics. Instead, it views all mathematical notation as strange glyphs, and so it generates its own strange glyphs to mimic the real thing. It would interesting to see how many people actually notice this.
<--nit picker (and one who also used to be a big fan of p-values)
Briggs, I like to nag you because I admire your love of ideas and your high-level of critical thinking. That being said, "what say you" to this proposition:
"There is an existing probability of distilled water boiling when brought to 100 degrees Celsius while at the atmospheric pressure associated with being at sea level. There is a second probability -- also one which "exists" -- of that same water boiling at 50 degrees Celsius at that same atmospheric pressure."
Did I just lie, or did I just tell the truth?
Does water really have this "natural propensity to boil under heat" that I claim that it does?
I’d be interested in the online class… partly because I almost failed a class on probability at a service academy 50 years ago. Had not the prof, due to his evident frustration of my density, basically given me the problems on the final during one of our many extra tutoring sessions I would have failed.
I believe the “credentialed class” use “studies” based on probability to manipulate the masses, just like the FDA and CDC use “studies” to manipulate the masses.
I ALWAYS learn something from your posts.
Thanks.
Am thinking about it.