Kenneth G from CA posted over 2 years ago:

Interesting article, but how can one know where the market is during a period of high volatility? Beginning, middle, or end? I would like to see the average duration of high volatility along with the standard deviation. Also, what factors are correlated with high volatility?

JOHN L from NJ posted over 2 years ago:

Another data mining based article that found a small negative correlation between volatility and returns. The observation that higher returns follow periods of high volatility isn't helpful for a long term investor. Obviously after all the volatility during the bottoming period of a bear market; returns are higher as high return early bull markets follow bear markets. This whole article is less useful than Warren Buffet's "Be fearful when everyone is greedy and be greedy when everyone is fearful".

BARRY J from TX posted over 2 years ago:

The length of this comment should warn you to skip it. Please forgive me, but I needed some mental activity to take my mind off the potentially apocalyptic events since October 6. Oddly, both sides share the same shibboleth, “Peace” and share a common fatal flaw, generationally ingrained fear of each other. I wrote this as a sorbet for my brain. Back to the article. Every day from 9:30 AM to 4;00 PM EST hundreds of millions of stocks are bought and sold instantaneously. If we aggregate these millions of mini-micro dots, they begin to look like a pattern, with hills and valleys, and as Mr. Israelsen does in his graphs, these dots can be ‘smoothed” out into lines using standard statistical techniques which our brains are conditioned to interpret in the context of an article as data or information. That’s the curse of backtesting. It can "prove" anything you want to if you torture the data long enough so you can present it in some form so our “lying eyes” can tell our brain to convert it into the first coherent story you want to believe. As a small boy, I was taught this small blessing, Always consider the source of what you think you know what you know. This is what my brain tells me I think I know I know. Different people see stock prices differently for many reasons. Prices move up or down for these real-world and “mysterious” reasons. Over time, some prophets/demagogues have stepped forward and claimed they know what the factors that move markets are. For example, in 1901, Charles Dow, of Dow Jones and WSJ fame, said it was “wave” patterns. In 1934, Ben Graham said it was accounting patterns he called “value.” In 1952, Harry Markowitz said it was “variation” patterns, and, contemporaneously, Paul Samuelson discovered that a Frenchman named Bachelier had described variation as “random” motion back in 1900. In 1963, Benoit Mandelbrot argued that daily price changes might be better explained and predicted using the math of power laws rather than Markowitz’s normal distributions because the statistics of normal distributions is not the appropriate tool for analyzing “fractal” events. In 1968, this became the Holy Grail of the institutional investor industry. Mr. Israelsen takes up this “un-Holy” Quest using data oddly enough beginning just two years after this event. We can characterize the modern era from 1970 as an “evangelistic” period where many new ersatz John The Baptist’ arose from the financial desert. Mr. Israelsen is an acolyte tending the alter in this long procession of academic wonderment about what “periodic” rhythms of price movements portend that can be statistically reduced into a possibly predictive ”formula” … and acted upon for profit by “unsuspicious” brains. In 2023, 53.5 years (or 619 months) later, in this article, Mr. Israelsen tells us the “Big IT” you may want to bet on is the degree of variation in variation SPX SD trends, kind of a second derivative variation of the SD of the variation of the SD argument. Mr. Israelsen’s conceit relies on chopping up a 53.5 year stream of trillions upon trillions of instantaneous data into 619 rolling 12-month chunks with the adroitness of a master Benihana sushi chef and runs these arbitrary chunked servings through a high-end Cuisinart statistical blender to produce homogenized smooth “standardized” frequency distributions that are pleasing to our lazy brains. The best metaphor for understanding what this process of torturing the data through all these mathematical gymnastics produces as “evidence” is the lesson from the denouement of “Raiders of the Lost Ark.” After reviewing all the “waves,” “random variations,” and “rolling 12-month periods,” the best personal lesson I derived is to remember that every time I buy or sell a stock, there is a counterparty who does the same, but, for reasons unknown to me, disagrees on what direction future price of this stock will take – for reason they feel is equally prescient and “informed” as mine, be they “fundamentalist” from “value,” “quality,” and “growth” sects) and “technicals” from “waves,” “chartist,” and “momentum” sects. Every time, we BOTH have a 50%/50% chance of being right. This cycle repeats itself millions of times every second, every day from 9:30 AM to 4 PM EST, and TRILLIONS of times in every “12-month rolling period.” The only Disciples that get rich are the ones selling “the Dream” of finding Salvation in the market by partaking of the latest Tree of Knowledge that they see as an “elixir” that can make you smarter than the other guy who thinks he is smarter than you are. In “Thinking, Fast and Slow (2011`) Nobelist Daniel Kahneman calls this “WYSIATI” (“what you see is all there is”), a term he popularized to describe the cognitive process that our brains are wired to believe that the information we have is all the relevant information there is. This becomes a problem because we tend to not look for what we don't see (evidence that would “falsify” our convictions). Our mental closets are full of get-rich fast and slow theories. “Taint so, Magee.” We are all welcome to follow the received wisdom of our choosing, but we must always be humble before the sacrificial altar of Mr. Market. This isn’t Kansas anymore, Dorothy. It isn’t Eden either. The denouement of Israelsen’s graphic dot saga is: If the SPX SD continues up in 4Q23, the forward 12-month (2024) SPX returns will likely be positive and likely large. If SPX SD goes down in 4Q23, the likelihood of a negative SPX return INCREASES. That sounds like 6-to-5 and pick ‘em bet on Wild Casino. Our consolation for taking on this 50/50 coin flip is we learned that “less volatility is not necessarily a good thing when considering the returns that follow periods of high volatility.” Less is more. More is less. Nothing confusing about that.

ROBERT A from NC posted over 2 years ago:

Calm down, Barry =). You do make some excellent points (as do Kenneth G and John L), but I like this article because it throws cold water on the flames of the myth that past volatility is a measure of risk. It also confirms what I already know: buying stocks in a bear market is a profitable undertaking. The more severe the bear, the greater the opportunity for profit in the following years.

BARRY J from TX posted over 2 years ago:

Thanks for the counseling, Dr. Robert. For therapy, I think I will go out and pound a 1.68-inch ball through time and space and try to bend it to my will. Based on a model from tracking past performance similar to Israelsen's, I estimate I will be as successful at that task as I am here. I am fairly certain that variations in my swing greatly influence the variations in my scoring outcomes. I actually liked Israelsen's optimistic message. Bilaterally, Israelsen concludes, *** "Should the [monthly] standard deviation of the S&P 500 continue to increase into late 2023, history would suggest that the forward 12-month return [for the SPX index ONLY since other indexes also exhibit variation] would be positive and likely large. Conversely, if standard deviation declines as we progress toward 2024, the likelihood of a negative return in the S&P 500 over the subsequent 12 months actually increases based on what has historically occurred." *** But as you said, it was the logic of his model that was problematic. Israelsen suggests that historical trends measured by the MONTHLY (an arbitrary time period) standard deviations in market INDEX variation (an arbitrary measure) are RELIABLE predictors of future market returns. Thus, it is volatility, not the health of the overall economy, nor the soundness of exchange-listed company business models, that we should look to predict market performance. Notice that Israelsen did NOT graph how daily periods of volatility foretell tomorrow's market performance, nor how the machinations of the Cboe VIX Volatility Index (VPCR), the Open Interest Participation, Open Interest Put/Call Ratios (OIPCR), Cboe Volume Put/Call Ratios (VPCR), or any of the many other measures of volatility the market generates daily, weekly, or annually. Thus, by process of elimination, it appears that the predictive powers of monthly volatility are based solely on some metaphysical covariance with the gravitational forces of a monthly lunar-based calendar cycle that make monthly variations movements the force that moves market index performance. This model is eerily similar to the tenets of Charles Dow's 1901 Wave Theory. I concede to Israelsen that his argument parallels Galileo Galilei’s argument to the Holy See when he defended his “heretical” theory that the Sun, not the Earth, is the center of the universe in 1633, "E pur si muove" ("And yet it moves.").

KENNETH R from WA posted over 2 years ago:

May I please have the contract to sell Barry printer ink and paper? :) Backward looking analysis to this degree is of limited use for an individual investor. While I appreciate the data salad for discussion, I’m not seeing actionable findings that might add enhanced utility to personal investment management. Impressive academic paper though.