Down goes the Dow

And over a long series of trials, being "right" (outcome closer to assigned probability) more often than not either identified a lucky guesser or somebody good at assigning realistic probabilities. A larger number of good guesses makes it more likely the latter than the former. But in a very large population, a small number of lucky guessers can maintain a lead in good outcomes for a long time. Those with a better track record over a long period of time are more likely smarter / better estimators, but some will get there as a matter of luck.

I'll ignore the weird condescending lecture on the definition of "100%, 50%, and 0%" but I'll note that it amuses me.

I'm not really sure what the rest of it is saying but it appears to be an attempt at explaining the definition of Bayesian statistics.

I'd appreciate further explanation on this please

Sorry racket, that's all I've got for today, and I don't have the energy for an argument. You can happily ignore everything I wrote if it pleases you. I was writing to amuse myself mainly. And now I must get back to work.

the idiot wrote:

Sorry racket, that's all I've got for today, and I don't have the energy for an argument. You can happily ignore everything I wrote if it pleases you. I was writing to amuse myself mainly. And now I must get back to work.

have fun

It's not really clear to me what the thesis the idiot posted above is about. It seems to suggest some sort of historic correlation and luck is...something or another?

But in case anyone is actually interested in how these things actually work, here's some good peer reviewed papers on generally accepted models for statistical analysis and model building for stock prices:

https://ro.uow.edu.au/cgi/viewcontent.cgi?article=1705&context=aabfj

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=424500

Anyways, NASDAQ to 10k today?

OK, I think I'm making some progress here.

Still a little unclear here, but this is, for the most part, false. Probability densities are not subjective. They are rigorously defined. It'd be akin to saying the real number line is in the eye of the beholder. I guess you could deconstruct it all the way down to a constructionist versus fundamentalist argument though (regarding the philosophical foundations of math that is).

Appropriateness of a fitted model can be tested through residual diagnostics (for stochastic processes).

You entire argument seems to suggest that all models are equal and none are better than any others becasue of some, i don't know, teleological stance you're taking??? Which is of course not true. I could do a basic linear regression on the SP500 over the past 30 years. Is it a model? Yes. Does it have predictive power? Yes. Do is it suck? Well considering the problem space is decidedly non-linear and highly stochastic, I'd say yeah it sucks a lot. Similarly, I can say with near certainty that in 1 year, the SP500 will be in a range of 0 and 10,000. Prediction? Yes. Valuable? Not even close.

the idiot wrote:

Ghost of Igloi wrote:At the February high what was the statistical probability that the S&P 500 would hit 2,180?

Great question. The chance, on a randomly selected day, of seeing a 33.9% or bigger drop 23 trading days later, is 0.4%.

Maybe that seems like a low number, but it means it happens, on average, every 250 trading days, or almost once a year. Which is not particularly uncommon.

However, that simplistic analysis is a little misleading, since all of the days when such a loss started occurred within the lengthy bear market of 2008. So, such a big drop happening so quickly is fairly rare. As are pandemics I suppose...

“Maybe that seems like a low number, but it means it happens, on average, every 250 trading days, or almost once a year. Which is not particularly uncommon.“

But that is not close to being true.

All eight indexes on our world watch list posted losses through June 9, 2020. The top performer is our own S&P 500 with a loss of 0.73%. Tokyo's Nikkei 225 is in second with a loss of 2.39% and in third is China's Shanghai with a loss of 3.08%. Coming in last is India's BSE SENSEX with a loss of 17.69%.

https://www.advisorperspectives.com/dshort/updates/2020/06/09/world-markets-update-june-9-2020?bt_ee=tUlHLCjjiZtXL60uGJpsTYn9hcfF2V8dZhQWErFzmJYURvPrVsXVcwnp6VhB80rp&bt_ts=1591791817541

Have been getting killed on my airline stocks last 2 days after a wonderful Monday.

Offset somewhat by Tesla stock continuing its spectacular rise.

Bought at $590 a few months back, now over $1,000.

Ghost of Igloi wrote:

the idiot wrote:

Great question. The chance, on a randomly selected day, of seeing a 33.9% or bigger drop 23 trading days later, is 0.4%.

Maybe that seems like a low number, but it means it happens, on average, every 250 trading days, or almost once a year. Which is not particularly uncommon.

However, that simplistic analysis is a little misleading, since all of the days when such a loss started occurred within the lengthy bear market of 2008. So, such a big drop happening so quickly is fairly rare. As are pandemics I suppose...

“Maybe that seems like a low number, but it means it happens, on average, every 250 trading days, or almost once a year. Which is not particularly uncommon.“

But that is not close to being true.

Igy, perhaps I could interest you in a concept called "Levy flight." It's basically a heavy tailed distribution where things tend to be business as usual, but when changes do occur, they're often severe (at least compared to the business as usual). Wikipedia has a good article on it and a graph that displays the concept well.

Portia wrote:

All eight indexes on our world watch list posted losses through June 9, 2020. The top performer is our own S&P 500 with a loss of 0.73%. Tokyo's Nikkei 225 is in second with a loss of 2.39% and in third is China's Shanghai with a loss of 3.08%. Coming in last is India's BSE SENSEX with a loss of 17.69%.

https://www.advisorperspectives.com/dshort/updates/2020/06/09/world-markets-update-june-9-2020?bt_ee=tUlHLCjjiZtXL60uGJpsTYn9hcfF2V8dZhQWErFzmJYURvPrVsXVcwnp6VhB80rp&bt_ts=1591791817541

Top performer US of A! Plus the NASDAQ is on fire

Ghost of Igloi wrote:...it means it happens, on average, every 250 trading days, or almost once a year. Which is not particularly uncommon.“

But that is not close to being true.

Maybe you read too quickly. That condition (33.9% drop over 23 trading days) occurred following 68 different days between April and September 2008. So from one point of view, it only happened "once" since 1950, over that much longer, deeper dive.

Beyond your statistical analysis, which I will defer to you, the event itself had few if any similarities to the past when you consider depth and time.

Racket wrote:

Ghost of Igloi wrote:

“Maybe that seems like a low number, but it means it happens, on average, every 250 trading days, or almost once a year. Which is not particularly uncommon.“

But that is not close to being true.

Igy, perhaps I could interest you in a concept called "Levy flight." It's basically a heavy tailed distribution where things tend to be business as usual, but when changes do occur, they're often severe (at least compared to the business as usual). Wikipedia has a good article on it and a graph that displays the concept well.

https://en.wikipedia.org/wiki/L%C3%A9vy_flight

Interesting. Fasanara Capital has uses a model where market structure becomes increasingly fragile via sameness, creating an environment that is viewed as stable, yet increases potential for large tail risks. Think index fund SPY, where there is no security selection, while investors assume both stability and liquidity for 500 positions. Central banks create an illusion of stability, but their actions manipulate market function, more through boosting confidence of investors, than actual real benefit.

Racket wrote:

Portia wrote:

All eight indexes on our world watch list posted losses through June 9, 2020. The top performer is our own S&P 500 with a loss of 0.73%. Tokyo's Nikkei 225 is in second with a loss of 2.39% and in third is China's Shanghai with a loss of 3.08%. Coming in last is India's BSE SENSEX with a loss of 17.69%.

https://www.advisorperspectives.com/dshort/updates/2020/06/09/world-markets-update-june-9-2020?bt_ee=tUlHLCjjiZtXL60uGJpsTYn9hcfF2V8dZhQWErFzmJYURvPrVsXVcwnp6VhB80rp&bt_ts=1591791817541

My stock portfolio is almost at exactly the same point it was at the end of 2019. That’s amazing considering all that has happened.

Top performer US of A! Plus the NASDAQ is on fire

Looks like a K5 post with that poor use of the quote function. Let’s try that again.

Racket wrote:

Portia wrote:

All eight indexes on our world watch list posted losses through June 9, 2020. The top performer is our own S&P 500 with a loss of 0.73%. Tokyo's Nikkei 225 is in second with a loss of 2.39% and in third is China's Shanghai with a loss of 3.08%. Coming in last is India's BSE SENSEX with a loss of 17.69%.

https://www.advisorperspectives.com/dshort/updates/2020/06/09/world-markets-update-june-9-2020?bt_ee=tUlHLCjjiZtXL60uGJpsTYn9hcfF2V8dZhQWErFzmJYURvPrVsXVcwnp6VhB80rp&bt_ts=1591791817541

Top performer US of A! Plus the NASDAQ is on fire

My stock portfolio is almost at exactly the same point it was at the end of 2019. That’s amazing considering all that has happened.

Fed saying now that rates are gonna be unchanged through 2022. Anyone still on the sidelines should just buy in now. DJIA is gonna be +30k by end of year easy

HTZ and CHK at $100 a piece. Better hurry now.

Interesting market response, NASDAQ up, Russell down. So, response at this point not uniformly positive.