The Finance Book Club

Its a bit of an unusual relationship. I receive emails from both Andrew and Tomas, and save them in the same folder, but its an uncomfortable fit.

I am not in a position to question the credibility of Tomas, however based on what he says I have not been tempted to fork over any dosh towards learning about his methods. As with anything, there is always an audience who will find something useful, perhaps even me. There is something to be said for automation - once you have your own rules down to a fine art. I don't have enough lifetimes for that to ever happen.

Edit - Andrew has a bunch of courses. I only just found about them now, despite having known about BST for I forget how many years. Wow. I have never doubted that Andrew is genuine in his intentions. A flag goes up whenever I read these words, though - "How even a beginner can become a high-performing trader.."

http://bettersystemtrader.com/courses/

Definite flag, of a certain shade of red. I am not one of those people who say that all income must be derived from a single source, and not everyone has the capital to make trading their main trade, even if they are good at it. However I am always skeptical of the sales-funnel shills who sell a course with no other guarantees. It's easy to prey on fools. Then again, without them, where would we be?

The 'Monday Effect' sounds like something someone invented in order to write an article about it, but if you believe the study cited it is real and it is certainly tradeable.

I have another word for it- momentum. I would guess there is also a Tuesday-Wednesday effect, a Wednesday-Thursday effect and a Thursday-Friday effect however they may not be as strong or as easily marked since Monday's trading often has no other interferences (from fundamental reports or news) to guide price in a different direction and so traders are necessarily 'anchored' to whatever happened in the previous trading day.

Does it exist in Forex/Crypto/Energy? I would guess so, but again, not as noticeably, and with less assistance from a strong trend bias. Still it makes you question the usual advice offered by sages. "Never hold a trade over the weekend." In fact, it seems that this is exactly what you should do, as long as you're trading with the prevailing momentum of the prior day.

If you haven't already please visit post 1 and vote in the book poll, as it will (maybe) help me break a 7-way tie.

The Monday thing - datamining. There are zillions of anomalies, limited only by the imagination. Yep, I've been there done that with heaps of stuff. Its another thing how to know how to correctly incorporate these things into a trading plan.

I know some of those books. There is one or two I'd recommend people actually reading. Some of Wilder is simply must know stuff, published in 1978. Its available on pdf, I forget how, and probably a breach of copyright.. Even the folk at metatrader can't code some fairly basic formulas correctly. Its little wonder many otherwise erudite folk say TA is useless when its not even coded right - this is nothing new with software products I guess.

Those pivot points that scalpers and day traders use? That's where they started.

I haven't read Miner, but apparently he goes into Gann and Elliot which can be complicated and dangerous. Bit like driving in traffic, even for the skilled. However -some optimising of oscillators sounds terrific. I know a bit about that and have my own way of doing using theoretical foundations rather than "optimising"in the traditional sense. Excel Solver. Its actually pretty much 90% of what I use. Took me years for the penny to drop on that.

Check out each one's Amazon reviews. That's a good start - the physics one appeals to me - "in 2008, sophisticated models fell into the hands of people who didn’t understand their purpose, and didn’t care. It was a catastrophic misuse of science. The solution, however, is not to give up on models; it’s to make them better." I reckon tat sounds like a good rant.

Ain't that the truth - I see misuse often, and then people turf it and say, well, that's rubbish, innit. Also see my second para here. I'm in the
"make them better" camp - what we are given is necessary but not sufficient. For example, Wilders volatility system is a rule of thumb thing, an approximation, for the Wiener process which is bread and butter for options traders.

The Excel Solver's use in trading would be a course/thread I'd be interested in attending. Econometrics in action.

It's a new week and I'm ending my self-enforced holiday as it's boring and I want to get back to work.
It's time to put up some pro numbers. It's time to learn something new and have fun in the process.
It's time to get things started on the Clemmo show tonight!

I have two clear front-runners and so I flipped a coin and will start with:

The Physics of Wall St. By Owen Weatherall. This will be followed by the other book to get two whole votes, Robert Miner's High Probability Trading Strategies.
Then I'll go in the order that the books were read, Penfold, Langer, O'Keefe, Cooper (still finishing) and DuToit (which I have yet to read). I'll do the others later, possibly in the order that they receive votes, if any.
Here we go!

The Physics of Wall Street - James Owen Weatherall

Author
Weatherall is a physicist, philosopher and mathematician with graduate degrees from Harvard, the Stevens Institute and UC Irvine where he is only an assistant professor of logic and philosophy of science. What do you need to do to get a full professorship these days???

Intro - of Quants and other demons

1. The book assures us that Jim Simons of Renaissance Technologies is the greatest money manager of all time, not Buffett or Soros
2. Simons is co inventor of the Chern-Simons 3 form - a string theory component
   1. For a better overview of Simon’s career and Renaissance, watch the long interview he gives Brady Haran on Numberphile

   
   
   

3. Together with James Ax they created the ‘Medallion’ fund named after the math prizes
4. In a decade the fund earned 2479%, blowing every other hedge fund out of the water
   1. And now? How is it doing now?
   2. Compare to Soros’ Quantum fund of 1710% (note they both owe something to physics in their naming/origins)
   3. Since the book was written Medallion has averaged 40% per year after fees which are 2x industry standard
   4. Compare to Berkshire Hathaway which averaged 20% per year and is not doing as well in recent years - but BH has the advantage of being transparent about its figures

   
   
   

5. Renaissance does not employ anyone from wall street. It hires only math and physics PhDs from the best schools. Isidore Singer calls it the best math/physics department in the world.
6. Hedge funds are supposed to work by creating ‘counterbalanced portfolios’
   1. Buy one asset
   2. Sell another asset as insurance
   3. The book goes on to define derivatives and futures contracts but you learned that in Schiller’s course or you can look it up on Investopedia

   
   
   

7. Despite their brains and best efforts on Aug 6 2007 the hedge funds lost millions or even billions
   1. The hedges didn’t go up when they should have; there was no safety net
   2. Goldman Sachs stopped the bleeding with $3B to stabilize their secret global alpha fund
   3. Recall Newton “I can can calculate the movements of stars, but not the madness of men”
   4. And also recall what Taleb said about quants and the futility of smarts against randomness
   5. How then to explain Simons?
      
      I have my own theory but I’m not sure if I should share it here. Ah what the hell.
      1. There doesn’t seem to be a flood of ‘Renaissance Technologies’ / James Simons clones making the market purely efficient based on neural nets and AI. Why is this?
      2. I’ve pointed to the fact that the market likely isn’t deterministic - it can alter as alteration finds. Also, the fact that smart analytical people have trouble with uncertainty. If a solution can’t be found by crunching mountains of data and analyzing it with a clever algo, then there simply isn’t a ‘solution’ (for them). Contrast this with a (sensible) gambler’s approach which is to treat the situation as mostly random, find favorable odds when possible, and avoid blowing up for as long as possible. This isn’t a satisfactory solution for desperados or quants, who both want to crush it. And in the end, how many poker players, even among the most successful, end up retiring rich? It seems maybe, fewer than you’d think. The game gets harder to play over time, due to competition. Adversity (and uncertainty?) is not something you can overcome with Math.
      3. However Simons has something that other firms don’t. A reputation for logic and mathematical brilliance. He hires only the smartest people who aren’t tainted by the technical analysis diseases of other financial people.
      4. He uses first rate computing and AI to find temporary patterns and exploits them before they are noticed by the rest of the market
      5. However this is a trick that you can only keep up for so long before everyone starts to copy it and the benefits dry up.
      6. This is one reason Simons and Renaissance are so secretive about what they do, enforcing their employees to what amounts to basically a lifetime vow of silence. Because once you even have an idea of what they’re up to, the patterns become obvious and the edge vanishes.
      7. That’s why I’m very curious to know how RT is doing in the 2020s. My guess is they look like mortals (along with the rest of the quants), but it doesn’t matter now as they have the investment monies, and the Fed-fuelled marketplace only goes up.

      
      
      

   
   
   

I must say the list of books you are reviewing aren't very mainstream, is it? Haven't heard of over half of them

You might be right but I take requests! What's your poison?

Chapter 1 - Primordial Seeds

1. This book is the story of physics in Wall St.
2. It is mostly made up of stories, which makes it a relatively interesting read

    

   1. This makes it a bit short on specific trading advice, sadly, however as we’ve seen so far, books that deal mainly in specific trading advice are usually terrible.

3. Louis Bachelier
   

    

   1. Gets a job at the Paris Bourse to support his surviving family ca. 1892
   2. An orphan who survived his military service
   3. Determined to become a mathematician or physicist
   4. Family business was sold to support his efforts
   5. Studies math at night at the University of Paris/ works at the Bourse by day
   6. Bachelier saw all market activity as ‘just an elaborate game of chance’
   7. Bachelier essentially invents the mathematics of finance, but he doesn’t get the fame he deserves until later

4. Paul Samuelson
   

    

   1. MIT economist ca. 1955
   2. Thought he was inventing mathematical finance
   3. Weiner processes, Kolmogorov equations, Doob Martingales - Bachelier had got there first
   4. Samuelson is shocked to have never heard of Bachelier

5. Leonard ‘Jimmie’ Savage
   

    

   1. Professor of Statistics at University of Chicago
   2. Writing a textbook on probability theory and statistics
   3. He was the one who stumbled on a 1914 textbook that had chapters dedicated to speculation and Bachelier’s findings
   4. He sends out postcards to his colleagues see if anyone has heard of this Bachelier character which is how Samuelson hears about him

6. Gerolamo Cardano
   

    

   1. 16th century Italian polymath (you can see how the book jumps back and forth in time which I find less entertaining in the Dr. Who sense, and more annoying in the Sliders sense)
   2. One of the most acclaimed physicians of his day
   3. He wrote essays on topics ranging from medicine, to math, to mysticism
   4. Loved gambling and used his mathematical discoveries to help him with cards and dice
   5. However his discoveries were not shared (naturally) until his papers were found

7. Chevalier de Méré
   

    

   1. Not actually a knight (chevalier)
   2. Early gambling revolved around the notion of odds and payoffs where odds were calculated using guesswork and intuition
   3. Intuits that rolling a die four times is likely to produce a six slightly more than half the time but cannot understand why rolling 24 dice produces worse chances of getting two sixes on any throw
   4. With this less than 1% edge de Méré is still able to make a lot of money gambling

8. Blaise Pascale
   

    

   1. A justly famous renaissance genius
   2. Not a gambler, but ironically remembered mostly for his wager about the existence of God
   3. de Méré meets Pascal at a Paris salon and asks him about his dice mystery
   4. Pascal enlists the help of Pierre de Fermat to solve this problem

9. Fermat
   

    

   1. Also famous - mostly for his ‘last theorem’ finally proved by Brit mathematician Andrew Wiles in modern times
   2. Lawyer and polymath, fluent in six languages, had heard of Pascale through Mersenne’s salon (recognize Mersenne as the inventor of a method of creating prime numbers?)
   3. Over the course of 1654 during a series of letters they work out the solution to the problem and inadvertently invent the foundations of the modern theory of probability
   4. “They were able to show that de Méré’s first strategy was good because the chance that you would roll a 6 if you rolled a die four times was slightly better than 50% — more like 51.7747%. De Méré’s second strategy, though, wasn’t so great because the chance that you would roll a pair of 6s if you rolled two dice twenty-four times was only about 49.14%, less than 50%.
   5. Cardano had also discovered this but no one knew about it at the time

10. So what is probability?
    

     

    1. There’s no guarantee that Cardano or anyone else will get the rolls they want just because the probability states it will
    2. So what do they really tell us?
    3. There is actually some philosophy involved. Probability is a relationship between frequency of events and desired events

11. Jacob Bernoulli
    1. The relationship is shown by Bernoulli nearly half a century later before his death in 1705
    2. Bernoulli showed that if the probability of getting heads in a coin flip is 50% then the odds of getting perfectly even percentages of 50% for any flip increases the more times you flip the coin
    3. Invents the very important Law of Large Numbers which you will recall from Aronson’s work

1. Henri Poincare
   1. Mentors Bachelier on his dissertation at the University of Paris
   2. Poincaré is practically the Einstein of France - made contributions to math, astronomy, physics, and engineering
   3. Had a lot of practical experience working as the chief engineer of the French mining corps

1. Bachelier again
   1. His thesis’ basic idea is that Cardano, Pascal and Fermat’s ideas could be used to understand financial markets
   2. One can imagine the marketplace as a casino; nowadays it is common to do so
   3. Sadly for Bachelier (and cosmically ironic as well) he never had much luck
   4. His thesis invented mathematical finance but his contemporaries didn’t understand it
   5. Mathematics was inward-looking and obsessed with rigor at that time (even now??)
   6. Since proving theories of probability rigorously is notoriously difficult, even now with computers, Poincaré could not award Bachelier the highest honors for his thesis
   7. It was not perceived to have been a dissertation about mathematics itself
   8. Bachelier’s remaining days are a professional disappointment as at age 44 (!) he is again drafted into the army before he can take up a permanent position at the university

1. Brownian motion
   1. Titus Lucretius notices bits of dust floating in light and calls them ‘primordial bits’
   2. Einstein - developed a mathematical framework allowed him to describe the trajectories of particles
   3. Jean-Batptiste Perrin - develops an experimental method to track particles suspended in fluid, confirming Einstein, and helping confirm that atoms exist
   4. Robert Brown - Scottish botanist noted the random movement of pollen grains suspended in water and gets all the credit because he’s British at the height of the British empire
   5. Brownian motion is a random walk - a drunkard’s walk
   6. Einstein gets the math credit for Brownian motion but in fact Bachelier had also beaten him to the punch in 1900, but unlike Einstein his paper wasn’t about dust motes but about the random movements of stock prices
   7. Stock prices are much like the drunk man in a hotel hallway who has an even chance of moving one way and then the other; there is a probability that can be calculated that after a certain number of steps he will be in front of any given room number
   8. This probability follows a normal distribution, or a bell curve, according to Bachelier

All well and good but why would you ever assume markets move randomly?! In fact, it’s still a hotly contested issue and we’ll examine it next.

Looking forward to your musings. I look at the markets as taking a trip from NY to LA in the states, we know where the destination is, but have no idea how, or when we are going to get there.

It sounds like you drive the way I do. Does your wife hate it as much mine does?
I think it is a good metaphor for how markets move, except sometimes the market goes to San Diego instead of LA.

Wife: Yayy! We are going to LA! How long will it take?
Me: Two days, give or take.

Wife: Why are we going South? I thought LA is West?
Me: Less traffic.

Wife: What?
Me: If we stay out of the major arteries during rush hour, we will get there faster......

In the MidWest.

Wife: Are we there yet?
Me: No.

Wife: I have to pee.
Me: You have to hold it for about an hour.

Wife: I can't.
Me: Stops on side of road.

Mountain West.

Wife: Can we go to Vegas?
Me: Yes.

In the desert.

Wife: I have to pee.
Me: Pulls over again. Car is is under stress from the heat. She is blissfully unaware.

Vegas. Two hours from LA.

Wife: Can we do the strip?
Me: Yes.

Los Angeles.

Wife: It took almost three days!
Me: I love you.

This is why people refer to the market as "she"

Genuine laughs out loud! Thanks for that.

Why would you ever assume markets move randomly

1. Bachelier knew from his time at the Bourse that prices go up on good news and down on bad news
2. This makes the notion of random walks in market prices crazy on some level
3. Looking backward from any instant it is easy to point to news to explain how the market moved (and when news conflicts, just pick the news that explains price movement best)
4. Bachelier was interested in the the probabilities of future prices - where you don’t know what the news is going to be
5. Bachelier believed that any predictable influences would already be reflected in the price
6. If you know something now that would make the stock go up or down, you logically will buy or sell based on that info
   1. Whatever is left over, reasoned Bachelier, logically makes future prices random (or at least unknowable)

   
   

7. This is the efficient market hypothesis (EMH) (you might recall it from Aronson or read post #69 of this thread) markets always reflect the value of a thing being traded because they incorporate all available information
8. Eugene Fama rediscovered this theory in 1965 to ‘great fanfare’
9. Of course the strong version of this theory is fishy at best - as my notes in that post show
10. Weatherall makes the analogy of high school physics problems taking place, usually, in a world without gravity and friction, for simplicity’s sake
    1. The EMH is a good starting point
    2. Bachelier left it for future generations to decide when it fails and come up with new ways to understand the market when it does

    
    

11. Paul Cootner
    1. 1964 - The Random Character of Stock Market Prices included an English translation of Bachelier’s thesis which gives the random walk theory’s credit to Bachelier

    
    

12. Samuelson - Foundations of Economic Analysis was groundbreaking where Bachelier could not be because economics as a discipline had matured into a mathematical discipline
13. Other influencers - EB Wilson, JW Gibbs, Irving Fisher, Jan Tinbergen (who famously argued that a company would become less productive if the income of the highest-paid employee was more than five times the income of the lowest-paid employee - completely forgotten now)
    
14. A ‘model’ like the random walk model is a simplified picture of how a physical process or system works

Fair Bets and Options

1. The random walk wasn’t Bachelier’s punchline - he wanted to find a way to properly price options
2. Weatherall explains what an option is which I’ll skip as you can look it up
3. Bachelier’s answer is based on the concept of a ‘fair bet’
   1. A bet is fair in probability theory if the average outcome involved for both people is zero
   2. Over many outcomes both players should break even
   3. An unfair bet favours one player

4. An option is a type of bet (in fact all trades are a type of bet)
   1. Bachelier reasoned the fair price for an option is the price where it becomes a ‘fair bet’

5. How much you gain or lose from an option is easy to calculate
   1. Difference in strike price from the market price for the underlying security
   2. With the random walk model we can calculate the probabilities that a given stock would exceed or fail to exceed the strike price in a given time window
   3. Using this we can calculate the fair price for an option

6. Markets are unpredictable because of randomness
   1. Predicting whether a given stock will go up or down is therefore hard
   2. Paradoxically though, the randomness of markets makes probabilistic determinations possible through the law of large numbers
   3. This is useless for direct speculation but it is useful for investors nevertheless

Paradoxically as markets are likely only ‘boundedly random’ this means you can’t just use probabilistic methods to profit from markets either. Diversification is not a perfect solution, and this -uncertainty- is necessary for the whole system to continue eluding the majority of participants, which in turn is necessary for a functioning market. This is all my own theorizing but I’d be surprised if someone hasn’t stated it more eloquently already.

1. Bachelier
   1. Offered no clear insight as to how to incorporate his options pricing model in a trading strategy which helps explain why it got less attention than his random walk model
   2. Options were exotic until the 1970s - illegal in the USA til 1960s
   3. A fiasco with Paul Lévy further damaged his reputation, unjustly
   4. Bachelier influenced Benoit Mandelbrot and MFM Osborne

Ok so if Bachelier got us part of the way, who did incorporate a rational options pricing model into a trading strategy?

Chapter 2 - Swimming Upstream

1. Maury Osborne

    

   1. Working at the Naval Research Lab he came up with an explanation for how insects can fly (drag + lift)
   2. Told his boss only 2 hours a day were necessary to complete his government tasks
   3. Osborne continued to work on his own projects at NRL

       

      1. General relativity
      2. Quantum mechanics
      3. Deep ocean currents

   4. Brownian motion in the stock market

       

      1. Written without knowledge of Bachelier
      2. In contrast to Bachelier, Osborne’s work was praised and encouraged because ‘physicists could do no wrong’ at that time in US history (cold war)

2. Weatherall goes on a long detour talking about Dupont and Nylon and uranium processing, but the point is that the barriers between applied physics/engineering and theoretical physics were crumbling
3. Osborne approached the problem of finding relationships in market data like an astronomer

    

   1. Build a model from experimental data
   2. He finds the numbers corresponding to prices move just like a collection of particles moving randomly in a fluid

4. Both Osoborne and Samuelson differ from Bachelier about the normal distribution of prices. Samuelson called it ‘absurd’.
   
   
   Attached Image (click to enlarge)

   
   
   1. Osborne’s shape had a hump and one long tail
   2. It’s what you get when prices are not normally distributed but the rate of return is
   3. RoR is the average % the price changes each instant
      

5. Investors don’t care about the absolute movement of stocks; they care about the percentage change

    

   1. Consider a stock that is $10 and goes up to $11. It made %10.
   2. Now consider a $100 stock that goes up to $101. Only 1%. Even though both rose by one dollar.
   3. Logarithms respect this size variation. The difference between log(10) and log(11) is equal to the difference between log(100) and log(110)
   4. “Statisticians would say that the logarithm of price has an “equal interval” property: the difference between the logarithms of two prices corresponds to the difference in psychological sensation of gain or loss corresponding to the two prices.”
   5. Note the importance of feeling based on a psych principle known as the Weber-Fechner law

       

      1. Small changes are noticeable at the start, but if the starting anchor is large, following small changes will not be noticed

6. Osborne was not influenced by authority

    

   1. He disagreed with Einstein about dark matter and in a series of now famous letters managed to show that Einstein’s book argument didn’t make sense and that he had no better ideas

7. Maurice Kendall

    

   1. 1953 showed that prices were as likely to go up as down

8. Osborne decides he needs to finish is PhD, at least to get a promotion at NRL

    

   1. His first dissertation on astronomy is rejected as unoriginal
   2. His dissertation on the stock market is rejected as not physics (i.e. too original, ahh academia)
   3. His third attempt was based on salmon spawning and was grudgingly accepted

9. There are apparently slow and fast fluctuations in salmon efficiency

    

   1. Markets also have these scales of fluctuations
   2. Details of how an exchange works or how traders interact can affect prices intraday (fast fluctuations)
   3. Business cycles, interest rates are more like slow fluctuations that affect price over longer terms
   4. Osborne went on to show that the same is true of ocean currents, fast fluctuations in water molecules could give rise to variations in large-scale systems like currents

10. Osborne spent the rest of his career trying to figure the ways that stock prices aren’t random

     

    1. Volume isn’t constant
    2. There are peaks in volume at the beginning and end of a trading day (exactly the opposite of a forex day), over the week and over the month - all slow fluctuations

11. Investors have limited attention spans (but robots do not)

     

    1. Investors get interested in a stock, make a lot of trades
    2. Then lose interest and volume declines

12. Including a volume model with the random walk model == the ‘extended Brownian motion model’

Next - the Osborne Paradox (my term)

The Osborne Paradox

1. “Osborne showed that if a stock went up a little bit, its next motion was much more likely to be a move back down than another move up. Likewise, if a stock went down, it was much more likely to go up in value in its next change.” (anti-momentum)
   1. At any given moment the market is more likely to reverse itself (form small channels)

2. “ If a stock moved in the same direction twice, it was much more likely to continue in that direction than if it had moved in a given direction only once.” (momentum - prices are not independent)
   1. Osborne argued that the infrastructure of the trading floor was responsible for this kind of non-randomness, and Osborne went on to suggest a model for how prices change that took this kind of behavior into account. (I’m not so sure but it hardly matters why price tends to remain in motion, what’s important is that it does.)
   2. Of course this has huge implications for the EMH

Huge implications as long as we can trust the methodology that Osborne uses. Can we?

this looks very interesting.

but it's not very clear (no offence!). "went up/down a little bit"... does it mean 1% ? 2% ?? 3% ??? and what about targets ? thanks.

That's an excellent question but the book does not specify. I am checking the notes in the appendix now to see if it will tell us or at least tell us where to look...

I have an opinion about this but I'll withhold it until I get the official story...