Probability Theory: The Logic of Science by ET Jaynes was the book that made statistics and probability theory finally make sense to me. Before reading this book, I was already familiar with Bayesian probability, having finished (among other books) The Signal and the Noise by Nate Silver, but Jaynes took everything to a whole new level. Probability is not an objective measure of likelihood—it is a subjective measure of incomplete information. This seems like a small shift in perspective, but it is much more profound than that, and you have to read the book to understand it.
The Goal: A Process of Ongoing Improvement by Eliyahu M Goldratt opened my eyes to process optimization. It is nominally about process optimization in a manufacturing plant, but so many of the lessons can be applied to other domains, such as software development. (And also about life in general, how to think about what matters and what doesn't.) It is a bit dated now; for example The Principles of Product Development Flow offers a much more systematic treatment of the topic. But I think The Goal remains the best intro to process and systems thinking.
> The apparent reduction in life expectancy for adults with diagnosed ADHD relative to the general population was 6.78 years (95% CI: 4.50 to 9.11) for males, and 8.64 years (95% CI: 6.55 to 10.91) for females.
It's not clear to me that there is a difference. It looks like the confidence intervals overlap.
You'd need to do a different analysis to decide whether there's a <5% chance of this data being generated by two identical distributions – but if we're not doing that, then we can't conclude that they're different.
This is just pure (somewhat informed) speculation.
ADHD is diagnosed roughly twice as often in men than women, and when diagnosed in women it is much more likely to be of primarily inattentive rather than hyperactive type compared to men. Many suspect ADHD may not actually be rarer in women, but just that the inattentive type is harder to diagnose, and more often never diagnosed- mostly because it is less disruptive to others in school settings.
I speculate that the women that are diagnosed are much more enriched for severe inattentive type ADHD, and I also speculate that inattentive type has a shorter life expectancy than hyperactive type. Just judging from myself, and other adults I know with ADHD: the hyperactive type ones tend to remain really physically active even into old age- unusually so, and the inattentive type tend to have a lot more depression, anxiety, and stress.
Isn't the combined type the most common, or has that type gone away? I find that people that are hyperactive tend to be more impulsive, and if I am not mistaken, impulsivity is the strongest correlated factor in determining longevity in humans. However, I could be wrong about the levels of impulsivity being different. That's just conjecture on my part.
I might be missing something but I think this may just be because women live longer on average. If we assume an equal increase in risk between men and women then women would automatically have a higher reduction in life expectancy.
Women typically live longer, so I was curious if this was a similarly proportional reduction in life expectancy for each group. Current life expectancy in the UK:
78.6 years for males, so 6.78 years is 8.6% of the average life expectancy
82.6 years for females, so 8.64 years is 10.4% of the average life expectancy
I think this is a relatively small difference in effect between the two, but I can't say for certain.
There is a difference in life expectancy in general so it would be remarkable if these are nonzero and the same. (I.e. exact same increase in risk of death at any pre-retirement age for both populations means higher change to life expectancy for women.)
Qubes has an excellent security model and should a top choice (if not _the_ top choice) for security-minded and technologically sophisticated users.
I used Qubes for a year or two, and then realized that my main use case was to isolated the browser, which to me was the greatest threat vector compared to everything else I use. Then I thought, if I just wanted a system with the browser isolated from my main Linux environment, wasn't that exactly what ChromeOS provided?
So I switched to ChromeOS and have stayed on it ever since.
The isolation in Qubes is much more reliable and flexible. I'm not even talking in Google's shady privacy practices. I'd never trust them with my OS or browser.
There is something wrong with some of those numbers.
For example, take 7-Zip Compression 22.01. The CPU Power Consumption Monitor chart states:
AmpereOne: Average 278.72W
EPYC: Average 311.64W
But the fine print under that same chart states:
AmpereOne: 6968J per run
EPYC: 14439J per run
By the Joules per run numbers, AmpereOne is far more power efficient than EPYC, requiring only less than half of the energy to complete a run.
In that case, how could the average power of EPYC to be only 11.8% higher than that of AmpereOne? For this benchmark EPYC is 14.2% faster than AmpereOne, and if the average power numbers are correct, the EPYC should have slightly lower Joules per run than AmpereOne.
That is not the only anomaly. For example, the CPU Power Consumption Monitor chart for John the Ripper 2023.03.14 also does not make sense.
The averages are not the same as the median values, I think this is where some of the problem comes from. The plots have quartiles with the boundaries shown as lines. The line showing the median value for the Ampre system is near the middle of the plot, but the median value for the AMD plot is far over to the right end of that plot, suggesting that many of the results were in a narrow range just above that value. This would skew the total average energy consumption way up, so we would see the difference shown in average Joules per run. This is probably not a good type of plot for this type of data, a scatter plot or line chart may be better.
Heres more:
* first chart has a n of 3. (Mythbusters' rocket car had n>3)
* the jouls you reference have candle charts showing way too much variance to make any conclusions.
Never-mind that these are all reduced to absurd levels, or biased.
My favorite was some site crapping on a SSD that only managed 3GiB/s for 100GiB of data, then dropped to 500meg or something. But, they didn't mention data transferred at all. Just speed vs time. Obviously pushing for that higher kickback on the ssd that costs 4x as much and uses 8x the power.
Real estate properties in Florida are increasingly at risk due to climate change.
Established insurers ran their numbers and noped out of Florida quickly.
New low-quality insurers came in and filled the vacuum.
Established rating agencies looked at the new low-quality insurers and came to the conclusion that they are trash.
New low-quality rating agencies came in and declared those new low-quality insurers to be a-ok.
Lenders, fearful of being left holding the bag (from climate losses that are likely to bring down the low-quality insurers), sold mortgages in Florida to Freddie Mac and Fannie May (i.e., the GSEs, which are financial institutions that purchase mortgages en masse and implicitly backed by the US Government).
Freddie Mac and Fannie May bought those mortgages because the properties are insured by (low-quality) insurers declared a-ok by (low-quality) rating agencies.
As is usually the case, the government will be left holding the bag.
The alternative is the NFIP [1], isn't it? The federal government was always going to be left holding the bag.
Meanwhile, responsibly run solvent programs like the California FAIR Plan for fire insurance have to fund it all from premiums and can survive major fires no problem.
Similar in a way to non-dischargable student loans. When lenders are not exposed to the real risk of their loans, someone always ends up holding the bag.
Mortgage loan rates in Florida probably ought to be in double-digits.
> Similar in a way to non-dischargable student loans. When lenders are not exposed to the real risk of their loans, someone always ends up holding the bag.
Students (people) are more likely to suffer from climate change and will have to fight it during their lifetime than those new low-quality insurers will have ever to, regardless of who hold the bags.
Let the fuckers hold the bag and give a fighting chance to the people rather than building our economy on the back of debt ridden slaves.
This is where the logic breaks down - lenders are selling these mortgages irrespective of any other conditions.
The only loans that aren’t sold are unsaleable investor and boutique loans. These wouldn’t be offered to normal borrowers in the first instance (since the loans of a normal borrower would be targeted for resale)
What I am missing is how do you profit from this. Many of these new entries have to know that the big insurance companies left for a good reason. That the smaller entries cannot absorb the risk they will be facing.
So, what’s the play? I have to assume there is some smart money who has identified a way of discharging the liabilities while still collecting pay, but how does it work?
I have no idea about these circumstances, but speaking generally...
It is really easy to make money from an irresponsible insurance company if customers have confidence in you. You take in a large amount of money from people insuring their stuff, invest it, make lots of money in the stock market, distribute the money back to the company owners as profit. Good times.
Then the highly predictable crisis that "nobody" predicted happens, the insurance company goes bankrupt, the customers get no payouts and the company owners don't really lose anything because they got their profits out years ago. Sucks to be a customer.
Under modern theory there is a Phase 3 where the government steps in to pay the insurance company extra money and keep them ticking over. It is an optional step and depends on political connections.
The basic idea is that insurance companies are paid to assume risk. But if, when the risk materialises, it turns out that the risk was actually held by the customers or government then it is a bit like the insurance company was making free money in the intervening period.
The goal is to underprice risk to drive out competition then grow to a point where you are too big to fail. All the while you collect bonuses as the leader of this scammy enterprise. You bank on events that happen every 50 or 100 years not happening before you grow so big you get bailed out when they do.
First, insurance boards by each state set the rules of the game, and one of the rules is to have an asset coverage ratio for the risk. Its effectively a sort of liquidity measure. Another one is minimum capital etc.
Both are traditionally paid by the original owners.
So, in effect, if the insurer goes belly up owners do lose some of the capital, and owners cannot loot the assets, either. But I agree that overall it is the customer who is most short changed because although they will get a partial payout in a crisis, it wont match their expectation - most of the missing funds will be gone as you have explained.
Thanks for this. I was anticipating some grand conspiracy, but nothing need be so complex. Just swoop in offering a product for a market who is required to buy. Pay yourself above market rates, and if you get lucky, you can maintain the system for years before ruin.
A mismatch in risk and term duration. In the worst case its short term cashflow, converted to profit ona regular basis. the huge tail risk is of a longer (likely) term. On a normal year reinsurance pays out expected losses, the primary insurer keeps their couple of percent margin, business continues. Eventually their optimistic/naive/malign actuarial numbers are shown up, reinsurance doesnt cover it, huge losses, bankruptcy, the profits are long gone and paid out, remaining share and debt holders are wiped out. Their insured customers are covered by the state after much bad pr, or just not covered.
Duh, I suppose this is obvious. Just take inflated pay as long as the good times last. I was trying to envision some grand huge payout, but methodical grind until apocalypse is fine too.
There are two reasons, one virtuous and one sinful:
* People hate saving money, money in our hand now must be spent now. Buying High and Selling Low is in our DNA, quite literally. Nobody wants to save a buck a day for a potential externality when they can spend that buck on something right here right now. Mandatory insurance forces people to save a buck (the monthly premium) for the potential externality, forcing people to Buy Low and Sell High.
* People want wealth, some moreso than others. The secret sauce? Being the middleman has the biggest revenue compared to effort. You are literally moving money from Person A to Person C, and you as Person B can take some of it. This is textbook rent seeking.
The Goal: A Process of Ongoing Improvement by Eliyahu M Goldratt opened my eyes to process optimization. It is nominally about process optimization in a manufacturing plant, but so many of the lessons can be applied to other domains, such as software development. (And also about life in general, how to think about what matters and what doesn't.) It is a bit dated now; for example The Principles of Product Development Flow offers a much more systematic treatment of the topic. But I think The Goal remains the best intro to process and systems thinking.