Johannes wrote:
Ghost of Igloi wrote:
From the Russians, probably a fake chart:
https://www.zerohedge.com/s3/files/inline-images/bfmF2F6.jpg?itok=4_V8wUF3It’s a real chart; just meaningless.
It really doesn't mean anything to me.
Johannes wrote:
Ghost of Igloi wrote:
From the Russians, probably a fake chart:
https://www.zerohedge.com/s3/files/inline-images/bfmF2F6.jpg?itok=4_V8wUF3It’s a real chart; just meaningless.
It really doesn't mean anything to me.
Hey, thanks Johnny. I was worried about the Russians.
Racket wrote:Cool, you mean like this guy?
https://en.wikipedia.org/wiki/Jim_Simons_(mathematician)
Maybe this guy found the pot of gold at the end of a predictive rainbow, but there's no way to test predictive skill after the fact. He may have amassed his fortune as a matter of skill, or it may have been simple luck.
If you take many million dart throwing monkeys (among which I count you, me, SP, agip, Igy, mas, ...), some of them will do very poorly, many of them will roughly match the overall market, and some will do very well. A tiny number will do extremely well, purely by chance.
The only way to test predictive skill is to register predictions beforehand and then measure their success over time. I know from my own experience (living life as an idiot) that I cannot outperform a dart throwing monkey. Which is precisely why I've chosen my screen name in this thread. Maybe DTM would have been a better nom-de-plume.
For those of you who believe you can out-perform a dart-throwing monkey, we can set up a simple forecasting experiment. It would take some time to watch it play out. You need enough forecast scenarios to be statistically significant, and enough players to distinguish between good forecasters and bad forecasters.
An interesting book on this topic that some readers may enjoy is Tetlock and Gardner's "Superforecasting." Also Judea Pearl's "The Book of Why" about establishing causation.
Yet you try to predict future market moves based upon statistical analysis of past market behavior. Seems loosely correlated to me.
Ghost of Igloi wrote:
Yet you try to predict future market moves based upon statistical analysis of past market behavior. Seems loosely correlated to me.
Igy, I think if you read back the tape, anything I have presented as forward-looking would be considered a forecast, not a prediction, per se. Sometimes I share guesses, which are also not predictions (to my way of thinking). The "statistical analysis" I've shared in the past is a first-order, super crude, naive look at past behaviour of the market. Or in other words, a representation of some simple metrics drawn from the set of past daily changes. The forward-looking projections / forecasts have been simple extrapolations based on past activity, presented with (very) broad ranges. Those broad ranges, which serve to make these projections of essentially near-zero value, have been one target of (fair) criticism on here in the past.
“Forecast” and “prediction” are synonyms.
Ghost of Igloi wrote:Yet you try to predict future market moves based upon statistical analysis of past market behavior. ...
Expressed differently (hopefully more clearly), I've given naive probabilities for the market to be within certain ranges at different time horizons, based on the simple (naive) belief that past behaviour (in terms of daily fluctuations) is a good first estimate of future behaviour (in the same terms).
Maybe I'll update that "forecast" in the next couple of days. My *belief* (the probability as interpreted from my point of view) is that the SP500 will never again reaching 1500, at any time in the future. Or that the probability (as reflected in my belief) is vanishingly close to zero.
But I won't attempt to "predict" what the market will do today, or where it will be in three weeks (except within a range with accompanying confidence bounds, perhaps).
George Corwin wrote:“Forecast” and “prediction” are synonyms.
Well, in laymans' terms, sure. In my line of work, they mean different things. To me anyway.
Semantics I suppose. Sometimes language is important. Often nuance is hard to get across in the absence of body language or eye contact.
the idiot wrote:
Racket wrote:Cool, you mean like this guy?
https://en.wikipedia.org/wiki/Jim_Simons_(mathematician)Maybe this guy found the pot of gold at the end of a predictive rainbow, but there's no way to test predictive skill after the fact. He may have amassed his fortune as a matter of skill, or it may have been simple luck.
He amassed his fortune by looking for non-stochastic events in the market and exploiting them in the 80s/90s. At this point, the fund kinda seems like a super secret Ponzi scheme.
Anyways, I've moved on from this as FB marketplace is providing serious entertainment. You wouldn't believe the stuff boomers are selling on here for prices all up and down the scale.
the idiot wrote:
Expressed differently (hopefully more clearly), I've given naive probabilities for the market to be within certain ranges at different time horizons, based on the simple (naive) belief that past behaviour (in terms of daily fluctuations) is a good first estimate of future behaviour (in the same terms).
Maybe I'll update that "forecast" in the next couple of days. My *belief* (the probability as interpreted from my point of view) is that the SP500 will never again reaching 1500, at any time in the future. Or that the probability (as reflected in my belief) is vanishingly close to zero.
But I won't attempt to "predict" what the market will do today, or where it will be in three weeks (except within a range with accompanying confidence bounds, perhaps).
The thing is: giving a forecast out of past behaviour could be almost relevant. But it still is all a matter of perspective.
My perspective - and I believe Ghost of Igloi's point is pretty similar - is that stock values (pease note that I don't say "price" here) should somehow be related to the bottom line results of both the present and expected future of the company, maybe corrected by some sort of confidence interval for their expected future results due to risk associated with the evolution of the context in which the company is doing business and/or strategic decisions made by the management. In making a buying (or selling) stock decision, you might want to assess if the price is higher or lower than your perceived value. This is basically called playing the fundamentals. That's in my opinion the way to do buy and hold. Not trying to outsmart and/or time the market, just assessing if the investment in the company is justified by last year, current year, next 3 years results expectations (some pevision is usually available for that), and even what you believe could be major tendencies in the next 10 years (for instance, I would be vary wary of a company exclusively doing mail distribution if they don't have a strategy for adapting their core model to the evolution of this market).
Now that this introduction have been made, to be clear, with prices as they are today 2020 price earning ratios are absolute garbage for most stocks, and 2021 are not a better outlook than they where before the crisis. In my opinion, most companies are currently valued either according to my best case scenario, or completely outside of it. And my worst case scenario would have some large to very large company simply going bankrupt.
Now compare to 2008-2009: PER did never went as high as they are now during that crisis, and by a very long shot.
IMO GoI might be a bit extreme in his 1500 prediction, but that could still be in the range.
expert of nothing wrote:My perspective - and I believe Ghost of Igloi's point is pretty similar - is that stock values (please note that I don't say "price" here) should somehow be related to the bottom line results of both the present and expected future of the company...
I can definitely appreciate that this is the conventional wisdom, and SHOULD be correct.
My own observation, though, is that the success rate for expectations (predictions, forecasts, projections) stated in advance seems to show little or no correlation to the experience / knowledge / expertise of the one stating them, nor the evident logical strength of rationale supporting their convictions. Market moves can only ever be "explained" (imho) after the fact. Or in other words, people can show evident skill hindcasting market moves, but rarely demonstrate consistent "skill," over the long term, looking ahead.
Which isn't to say successful investors don't emerge, of course they do.
A great example of failure to predict market future based on fundamentals would be Igy calling for a crash since ~ 2015. It *should* happen, but hasn't (so far).
All eight indexes on our world watch list posted losses through May 28, 2020. The top performer is our own S&P 500 with a loss of 6.22%. China's Shanghai is in second with a loss of 6.68% and in third is Tokyo's Nikkei 225 with a loss of 7.36%. Coming in last is India's BSE SENSEX with a loss of 21.95%.
My biggest single frustration with the current market is that so much of the movement takes place when the market is closed. Case in point. I own CGC. It closed at 21.72 yesterday and opened at 17.04 today on poor earnings. I liked it at 21 so I tripled my small position at 17.95.
OK.
fisky wrote:
My biggest single frustration with the current market is that so much of the movement takes place when the market is closed. Case in point. I own CGC. It closed at 21.72 yesterday and opened at 17.04 today on poor earnings. I liked it at 21 so I tripled my small position at 17.95.
Been this way since about the 1985
Racket wrote:
fisky wrote:
My biggest single frustration with the current market is that so much of the movement takes place when the market is closed. Case in point. I own CGC. It closed at 21.72 yesterday and opened at 17.04 today on poor earnings. I liked it at 21 so I tripled my small position at 17.95.
Been this way since about the 1985
i see they announced earnings before market open today. Did they announce earnings during pre-market trading? If so, that's par for the course.
That's a sector (pot stocks) i dabbled in a year or two ago, but i have steered clear ever since.
Good luck and with that area, things change quickly. Hopefully for the better.
Earnings Scorecard: For Q1 2020 (with 97% of the companies in the S&P 500 reporting actual results), 64% of S&P 500 companies have reported a positive EPS surprise and 56% of S&P 500 companies have reported a positive revenue surprise. If 64% is the final percentage, it will mark the lowest percentage of positive EPS surprises since Q4 2012 (also 64%).
Earnie wrote:
Earnings Scorecard: For Q1 2020 (with 97% of the companies in the S&P 500 reporting actual results), 64% of S&P 500 companies have reported a positive EPS surprise and 56% of S&P 500 companies have reported a positive revenue surprise. If 64% is the final percentage, it will mark the lowest percentage of positive EPS surprises since Q4 2012 (also 64%).
Your post is a off color joke, Earnie. Yea Q1 GAAP EPS on the index is $12.51. Now that is a surprise on a quarter that was suppose to be three times higher. Wait until Q2 spin on under $10 EPS. Oh, that’s right a V-shaped recovery. :-)
https://www.wsj.com/articles/the-markets-are-wild-while-youre-asleep-11587466802seattle prattle wrote:
Racket wrote:
Been this way since about the 1985
i see they announced earnings before market open today. Did they announce earnings during pre-market trading? If so, that's par for the course.
That's a sector (pot stocks) i dabbled in a year or two ago, but i have steered clear ever since.
Good luck and with that area, things change quickly. Hopefully for the better.