Discover and read the best of Twitter Threads about #rwri

Most recents (15)

My personal generalized thoughts and conclusions from #RWRI 12 @nntaleb

-Statistics is about the elimination of noise
-Decision making gets easier in uncertain conditions
-Skin in the game is a filtering mechanism
-Group behavior is not expressed by individual preferences
-Nature is scale invariant
-You create tail events simply by regulating the process - it’s an unnatural constraint
-It’s misconstrued that lowering variability actually lowers risk
-McDonalds exists as an insurance policy
-Everything non-linear is necessarily convex/concave
-Don’t theorize about what people want, just act on what they do
-Have optionality on the convexity of random events
-PTSD vs Post traumatic stress growth. One is talked about because it makes $$$
-Floor your losses
-Zero intelligence trading works when you have convexity bias
Read 13 tweets
“If you put a blueprint on anything that works organically you’ll f*ck it up.”

[Day 5] #RWRI 12
@nntaleb “Anyone who gives a zero probability is either a religious fanatic or an idiot.”
@nntaleb “Thou shall not buy a put from anyone who’s performance is correlated to that put.”
Read 6 tweets
“Just knowing that the tail is there is half the battle.”

[Day 4] #RWRI 12 … some nuggets of wisdom
@nntaleb “If you need a good reputation to survive, you’re f*cked.”
@nntaleb “Any system that needs confidence to function is about to go bust.”
Read 6 tweets
“Rule of thumb: you’re in a drawdown state 70 % of the time. And more than half the time in a major drawdown.”

[Day 4] #RWRI 12 … facts about 180 years of stock market history
@financequant [Drawdown history without the great depression] “…its’ like saying that instead of having 100 pounds of TNT blowing up you have 98 pounds of TNT blowing up.”
@financequant “If everyone is running for the exit does it really matter why?”
Read 4 tweets
Ok, the Russians were here, and I didn't understand a thing. Next up @trishankkarthik, who's claiming Quantum Supremacy isn't a racist thing. Let's see how this goes.

Taking an integrative non-segregationist view, he's explaining that all computers are basically the same. #RWRI
Ok, so, point is that some things are logically impossible. There is a perfect answer but it takes a shitton of time (which you don't have) to find it out... But if you're given an answer, much easier to figure out if it's right or not. #RWRI
Read 6 tweets
Robert @financequant kicking off #RWRI today with 180 years of drawdowns 📉

He is surprised how no one else has looked into this, especially that far back

There is this myopia, this attitude that looking that far back doesn’t have anything to do with us
As Robert says, the math is not very complicated

And he went into this fully expecting to find more stability, because we have all these wonderful tools now

What is a drawdown? If you lose 10%, you need to gain 11% to make back the money
What drawdowns are: two simple variables

How long and how deep
Read 19 tweets
“Sometimes the best way to reduce uncertainty is to do f*ckin’ nothin’.”

[Day 3, #RWRI 12] … some insights
@nntaleb “Backtesting is for morons. With backtesting you don’t capture the shadow mean; only the mean.”
@nntaleb “You backtest a strategy not a future trade.”
Read 4 tweets
1/n Just listened on way to #RWRI the @econtalker podcast with @mazzucatoM about the government role in innovation "investor of first resort". Will comment further after I read the book, but found many flaws in reasoning by Mazzucato:
@EconTalker @MazzucatoM 2/n Govt bureaucrats are not boring & lacking in innovative drive because they didn't study strategy in business schools (i.e. because of their education) as claimed by @mazzucatoM, rather boring people tend to work for government.
@EconTalker @MazzucatoM 3/n @mazzucatoM claimed that by attending Princeton high school she attended a "state school". Scale effect: it's a school for the residents of the commune, equivalent in its functioning to a private school.
Read 4 tweets
1/3 Order from randomness: @financequant demonstrates #complexity emerging. Check out what comes from this simple algorithm iterated... #RWRI
2/3 Voila! The goddamn Sierpinski triangle! #RWRI
3/3 I really suggest trying this out on your computer, to get the eery feeling of demonic possession. (@FredHasselman has R code here:…) #RWRI
Read 6 tweets
Lenin’s elder brother Alexander plotted to kill the Tsar Alexander III while he was a young student in Saint Petersburg. His plot was uncovered and he was executed. The Ulyanovs were shunned by Simbirsk ‘good society’. This explains Lenin’s quote the “personal is political”.
The world is quite small at the top: Alexander Ulyanov (Lenin's brother) studied chemistry with Mendeleyev, who was appalled when hearing about the execution. And Josef Pilsudski, who would go defeating Lenin in Poland some 30 years later was one of his conspirators.
Lenin read Chernyshevsky manual on how to become a revolutionary (What Is To Be Done?) five times during the summer he would be expelled from Kazan University. Inexplicably the book had not initially been censored and became an instant hit...
Read 75 tweets
Updates: So I haven’t been on here as much. I wanted to share some progress.
I was originally going to start a side business in 2020. I was merely planting seeds this year. Securing a promotion at work and scouting leads.
Then it hit me. My natural aptitude and some the ideas I’ve gotten from the #incerto series/ #RwRi crew can help. So on a whim I started with just 4 things. In no particular order. It starts with attitude. 1) Be patient. 2) Don’t lie (meaning if you don’t know say so)
3) Don’t be lazy. 4) Put maximum effort even in minimal side jobs. 5) Don’t complain if you get short changed, just learn, move on and create a black list. 6) Teach others around you so they can take on extra work loads. cc: @ember_sadat
Read 8 tweets
DataScience 101
#RWRI showed how elementary but nonlinear concepts like "correlation" are poorly understood by pros.

One needs to translate ρ into information.
See how ρ= ½ is much closer to 0 than to a ρ=1. Yuuge diff betw .9 & .99
Psychologists cluelss for 50 years.
2) The entropy/information theory approach to correlation is mutual information, similar to Kelly criterion:
How much can I bet on the value of Y knowing X? If Y certain (ρ=1), infinite. If ρ=½, tiny.

All IQ psychology is worthless.
Trying genetic maps: they are misinterpreted. "Linear correlation" is nonlinear to information.
But a trick for rescaling works when data is Gaussian.
cc @ArthurB
Read 4 tweets
1) I have always approached business from the asymptote to the middle. At #RWRI, this was called working from the fat tail -> middle. When you are buying media, best to start figuring outfraud and "impression inflation" first.
2) Impression inflation is a cousin 2 fraud. Ever click on content "30 celebrities caught in ..." Each celebrity pops a new slide. What is an ad impression on the 17th side actually worth? And what about ad impressions below the fold? So close to worthless-let's call it 0 value
3) Automatic, programattic ad buying will include these worthless impressions. The entire online media ecosystem is f***ed, which is why you need to create and own your own media. Stay tuned for real world marketing tactics.
Read 3 tweets
5th day #RWRI 11
5th day #RWRI 11
Machine learning
Read 3 tweets
Today #RWRI 10, we discuss:
How the "calibration" ranking by BS Vendor @juliagalef is not even wrong

1) In real life, you shb be calibrated in payoff space, not in frequency space. Someone right 80% abt Russian Roulette won't survive, option traders can be right ~1%
2) Which is the idea of #skininthegame as a filter. The ONLY filter.

Verbal predictions do not map to anything real, as they penalize those wrong most of the time in skewed payoffs and favor bureaucrats who optimize in frequency space.

Lookup the story In Fooled by Randomness about the trader who thinks the market is "more likely to go up", yet is very heavily short.
Read 5 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!