my journal 2

Status
Not open for further replies.
Contigo En La Distancia-Christina Aguilera

No existe un momento de día
en que pueda apartarme de ti;
el mundo parece distinto
cuando no estás junto a mí.

No hay bella melodía
en que no surjas tú,
ni yo quiero escucharla
si no la escuchas tú.

http://www.youtube.com/watch?v=9s4RgVrAG7Q

Es que te has convertido
en parte de mi alma,
ya nada me conforma
si no estás tú también.

Más allá de tus labios
del sol y las estrellas,
contigo en la distancia
amada mía estoy.

César Portillo de la Luz
 
mumford

watchable:
http://www.letmewatchthis.com/watch-146152-Mumford

Wrong: great movie. I will have to check out the director:
http://www.letmewatchthis.com/?&director=Lawrence Kasdan

One in a hundred, a rare good movie. Which basically means "a movie I really liked". Yeah, because they are rare. I almost always find something wrong with them.

Typically a movie I dislike is a movie with Tom Cruise, Brad Pitt, Nicole Kidman. These actors really did not make a single movie I ever fully liked. They're like a guarantee of watching a worthless movie.

The typical movie I like is for example a movie with one of those SNL comedians, or one with Jim Carrey, or Steve Buscemi, Harvey Keitel, Matt Dillon, Christina Ricci. These people are a guarantee of quality. This used to be very true with De Niro and Pacino, but it is no longer true, with the movies they made in the 1990s and 2000s.
 
Last edited:
Last edited:
Manhã de Carnaval

I can't get it out of my mind. Especially the girl. But also the song.

http://www.youtube.com/watch?v=22DEBZzi_vE

Manhã tão bonita manhã
De um dia feliz que chegou
O sol no céu surgiu
Em cada cor brilhou
Voltou o sonho então ao coração


Depois deste dia feliz
No sei se outro dia havera
E nossa manhã, tão bela afinal
Manhã de carnaval



Manhã de Carnaval
 
Some reasoning about in-sample and out-of-sample

I was talking to another trader and he said that, as far as trading systems, the future is rarely as good as the past.

I don't know much about statistics but I'd like to reason about this out loud.

I just want to reason about it. Not read any papers. Not do any tests. Let's see if I can get to the conclusion by a logical deduction.

If I don't, it is just because I am in a distracted state of mind, due to concerns about personal issues in my life (work, vito, neighbours, etc.).

Yeah, because I think there's no need for empirical evidence to establish this.

So, let's say that on tradestation I test a bunch of hypotheses for systems that I think should work, and I come up, out of 10 strategies tested, with 5 systems that work in the In-Sample.

By the way, we should clarify that we have three phases:
1) In-sample, where I look for hypotheses that work (with the help of optimization)
2) out-of-sample, where I verify that those hypotheses weren't just lucky combination of rules/parameters
3) real trading, where I trade those systems with real money

Now, what my trader friend said, and I think he might be right, is that the "future is rarely as good as the past". So this should also be true for the out-of-samples. If it is true, that is.

At this point maybe we should add what "good" means. It means profitable. And we are up against not just chance but also costs (commissions and spread). So no system, by randomness, will ever be "good", "profitable", because we are up against randomness PLUS costs.

So, as a rule, a random system will be unprofitable, for the simple fact that it will be a break-even system MINUS costs. As a rule, only a system with an edge will be profitable. With an edge large enough to cover costs and add some profit.

Now, what my friend is saying is that there will be a tendency, within those 5 profitable systems, to not perform as well in the out-sample: not necessarily to become unprofitable.

Let's say that for a second we go back to my ten strategies before any optimization.

The chance of them performing in the out-of-sample as well as in the in-sample, is identical. As long as I don't mess with the parameters and the rules based on what I learn, by optimizing (i.e. "data snooping"), my rules will have an equal chance of working in one sample (the in-sample) as well as in the other sample (the out-of-sample).

So what is critical here is the process of optimization: that is what makes my systems likely to not perform as well as they have.

One more remark: my friend may have been referring also to the markets changing in the future, but let's set that part aside. However that part alone means that even if you did everything correctly in terms of methodology used, your systems won't perform as well. They could even perform better, but they are likely to perform worse, because you always will have commissions against you, which will be overwhelming if your edge ceases to exist. And, according to my friend, that is usually the case. After a while, that edge disappears. And I trust him on this sad reality.

So, as a rule, the system will not perform as well in the tests, and we have established this, but this is not what I wanted. Because all this means is that the system will not work as well once it stops working.

What instead I wanted to reason about and find out is whether, due to our process of optimizing, we're causing the out-sample to be worse than the in-sample.

Here I can't forget about my experience. It is true that out of dozens of systems I created (about 100), only about half of them worked in the out-of-sample, which means "were profitable in the out-of-sample". It doesn't even mean that they were as profitable, but just that they were profitable (after fixed costs).

So this proves it is true.

And I suppose this is because not all things that happened in the past happened because they followed a pattern that they will follow again. Sometimes they happened randomly. Then I come, and, via optimization, I piece together those random acts of the market, and find a pattern to them. The problem is that I may find a pattern that was not the pattern the markets followed.

The more reasoning you do on your system BEFORE optimizing it and the less you optimize it, the more you're likely to have success in the out-of-sample. And viceversa. I have not tried them yet, but I think this is why maybe genetic algorithms may not be so useful, in that they find good brute-force combinations. They are nothing but smart brute-force optimizers. And what we need is not combinations but reasoning. The combinations are easy to find with the regular brute-force optimization.

One should not even use a genetic algorithm if he doesn't know the markets, because they will produce so many systems that of course some of them are also bound to work in the out-of-sample.

One thing is to produce 10 strategies and 5 of them work in the out-of-sample. Another thing is to produce 10 strategies and 1 of them works in the out-of-sample. In the latter case, you risk finding combinations that work in the out-of-sample by luck. Your ability as a system-creator should be measured by how many of your systems fail in the out-of-sample. In my case, 50% of systems fail, and I feel ok about this ratio.

So, recapitulating, we have established that systems we create are not likely to work as well as in the in-sample, because:

1) the markets change and have a tendency to reduce your edge (from a specific system)
2) the process of optimizing on a sample data set gets you acquainted with what happened, and the past is not likely to be repeated that way - even if it didn't have a tendency to reduce your edge.

But there's still something missing. Two things. What is the relationship between in-sample, out-of-sample and real trading, and how much are the systems going to be worse?

Let's start again.

1) I test my strategies on the in-sample
2) 50% of them are profitable in the out-of-sample. Profitable enough to be worth trading. But it is true: slightly less profitable. Let's say they are roughly two thirds as profitable, on average.

And so far my friend was totally right. Because he was proven right by the fact that half the strategies do not work, and the other half do not work as well, only 66% as well.

Now let's get to the last part: the part where I trade the strategies.

More time has gone by. We are now in the out-of-out-of-sample.

My edge is likely to have disappeared even more, because more time has gone by. However, in terms of sample, there should be no difference, or yes?

Let us completely set aside the time factor which increases the chance for the edge to disappear (due to other traders trading the same strategy - the usual "efficient markets" talk).

Yes, yes, yes! There is still another difference.

It is the same difference I mentioned for the genetic optimizers creating hundreds of systems.

If we create a large amount of systems, even using the out-of-sample methodology, we are bound to find a lucky brute-force random combination that not only works out of luck on the in-sample but also work out of luck on the out-of-sample. Such a combination could happen maybe... guesstimate: 5% of the time or less. But out of dozens of systems created, there will be some systems that were lucky in both in-sample and out-of-sample and those systems will fail in real trading.

That is why it is of major importance that we have a good ratio of systems successful on the out-of-sample vs the amount successful in the in-sample. The more systems we verify on the out-of-sample the more likely we are to find a lucky one that will be profitable by pure luck. Ideally, we should have a feeling for this, and sense when a system is not ready to be tested on the out-of-sample. In fact, 6 months ago, until the investors told me to use it, i totally disregarded this method, whereas today I consider the out-of-sample verification as something sacred, to not be abused, only to be done once, at the end of all testing.

The result of not using the out-of-sample in my previous tests is that I ended up automating a good 50% of unprofitable systems, and my forward-testing took the place of the out-of-sample, and I needed to wait a year to find out which systems really worked.

Recapitulating, I've established that:

1) i test my strategies on the in-sample
2) half of them work on the out-of-sample but only achieve 66% of the performance they had on the in-sample
3) due to time eroding the edge, and due to lucky out-of-sample combinations (one system every 20) the performance of my systems will further decrease and in some cases they will never be profitable.

All in all, given my skills and work, I can expect the performance of the systems I create and approve for real trading, to be on average 50% as what it was on the out-of-sample. Yes, because by the time the good systems reach the out-of-sample they decrease by 66%. By the time they reach real trading they decrease another 16%, due to edge being eroded and lucky systems that were let through.

That partly explains why today I am showing only a small part of the profit I was expecting. This is also due to a poor selection of systems to trade. I had picked some systems that were actually unprofitable in the out-of-sample, too, just because I "trusted" them.
 
Last edited:
pen stolen (by vito)

Damn. I came this morning and my pen was gone. I know exactly where I put it last Friday before leaving the office. And the only person who knows where I hide it is Vito. So Vito took it, 100% sure. It sucks to have to be at the office with such a disloyal person. I don't do the stuff he does, so he totally wins at this type of guerrilla. I can't start stealing stuff even if I were to do it to someone who does it to me. My education is stronger than my resentment towards him. Besides, I would not want to be suspected of stealing by anyone, even him.

However, I can retaliate on a smaller scale. For the last month, I haven't talked to him, but I've helped him when he asked me for help. Today I won't even help him at all. Maybe I will be like this for the future, too. The carrot is gone, since it's useless. It will be the stick, permanently. A stick he is not even afraid of. Maybe this is not so good, but I can't help someone who messes with my stuff. I will just say that I am busy. This is what he gets for starting to mess with my stuff again, even after being lectured by the bosses. I can't believe I am having to deal with such a scumbag at the office. I can't believe such a jerk got hired - yeah, because I think he already got hired. He fooled a lot of people but not me. From now on, no more help to him. He was the only one who knew about the pen. He's a very disloyal person. I really hate this guy.

In a way he is sick. This behaviour is sick. Almost everyone else would have stopped by now. I've never met someone so spiteful and at the same time so disloyal and fake - yeah, because he has the guts to ask me if I want him to bring me back a sandwich when he goes out for lunch. He actually pretends he is my friend. Then, when I am not there: scanner and keyboard flipped, cords messed with, envelopes spread out all over my table... and it's him for sure. But he's not going to drive me crazy, if that is what he's trying to do. He will just lose everything I can offer to him. He lost my friendliness, now he's losing my help. If he does even worse, I won't even say hi to him anymore. It's hard but it's an obvious choice: I can't just keep on being nice to someone like this.

94846d1288519357-my-journal-2-ugly.jpg


Mmh. I am having second thoughts about this. What if it wasn't him? Here's what I'll do. For today I will say that I am busy and I won't help him. That way if it's him, he knows why I am acting like this. If it's not him, he will just think I am busy.

The reason I am having second thoughts is that yes he did mess with my stuff for sure in the past. But he's never stolen anything... damn! Just as I said this, I looked again, and I found my pen still there, under my desk, but on the other side (I usually hide it on the right, but this time it was on the left).

All right, next time... he gets a bonus for this. I will always help him no matter what happens to my stuff. But I won't make friends with him ever, for he did flip my stuff in the past and all that.

pen found (it was never stolen)
 
Last edited:
things done, to do, and postponed

With a lot of problems behind me, and some projects postponed, here's what's behind and ahead.

Things done:
Vito: I've disciplined him and his new behavior (silence) is stable now
Part-time: within one month, I will be able to go home at 15.00.

Things postponed:
Buying ts-evolve by katz: I don't need it now to create more systems. I will buy it in a couple of months or so. But have to do it before they stop selling it, so within the next two or three months.

Things to do:
Funding IB account: I need to fund my IB account quickly and also get some money from the investors, so that I won't be at risk of gambling (if their money is in it, too, I won't gamble). Then we need to either trade it discretionary with their trading choices. Or we need to allow one of my systems (the least risky one) to trade it, so that IB will be happy and won't close my account for inactivity.

Getting ideas for new systems: Since I have a lot of free time and there'll be even more in one month (due to the increased part-time), and since I've realized that I have gotten very quick at testing and implementing new systems (in one weekend I can complete all the work required for one system or even two), I need to look around me for new ideas to test, the so-called "inefficiencies", term which I dislike. Here's what I'll do. Two things
1) I will look in the trading forums, like this one, for people saying what they do, and I will test what they do on tradestation.
2) I will ask non-traders for ideas on what the charts do and what people do in the street. My systems are all so simple that they can benefit from non-traders as much as they benefit from traders. A non-trader doesn't just have to look at a chart. I could also ask him "how can you predict, based on the past, where a pedestrian will cross the street?" and similar questions. Predicting the behavior of people is close to predicting the behavior of the markets. I will focus on this section of the things to do on my next post: "Getting ideas for new systems".
 
Last edited:
Getting ideas for new systems

Let us start gathering some links from this very forum, of journals where people say something about their strategy other than just "long here" and "short there". Also, I need journals of people saying they are profitable. It doesn't matter if I don't have the proof: those, like me, who say they are not profitable are certainly not profitable. So it's a start, to look for those who say or sound like they are profitable.

Gathering ideas from traders
ForexMorningTrade journal (trading at the London open - vendor)
Goldmine! (Andrews Pitchfork)
Alpha Markets - Simple Forex Trading (Hemal Pandya’s posts are clear and to the point - very promising thread. Only problem he is in fact a vendor)
Spreadbetting FX using TA
My Journal at a Prop house
Entry=Risk Exit=Reward
Vwap Engine

Gathering ideas from non-traders
For this I can start by asking myself the question as follows: if we were predicting the behaviour of a pedestrian, how would we go about calling the direction where he will go? Now, we cannot predict some direction that he's likely to take, because in the markets we have to predict something that is unlikely or else we cannot cover commissions. To do this I would have to observe pedestrians for a while. I should sit at a bar, identify someone walking down the street, and try to predict whether he will turn one way or the other. Then I should look at what reasoning is going on in my mind and if I can apply any of that reasoning to the markets. In the same way, I should ask the same type of question to others. Another good general question is "how well can the past predict the future?". In nature, this is too easy: in winter it's cold and in summer it's hot. I need to study a subject where the laws are not so certain and clear. Something like the markets. I have to make a lot of comparisons with the markets.

So here's what I will do in the next few weeks: question myself and others about the nature of predicting the future based on the past, on the logical reasoning involved in this. And study the journals listed above, to see if I can get any ideas for more trading systems.
 
Last edited:
expanding on previous post

continuing reasoning from previous post:
http://www.trade2win.com/boards/trading-journals/85510-my-journal-2-a-178.html#post1330742

I am going to start off from very afar and tackle the issue very broadly:
Predicting the future: it's becoming a science

...This exciting new field of science is known as 'threshold and pattern dynamics', because it involves understanding when a vital threshold (to a catastrophe or major change) is crossed - and recognising patterns in the things that are driving it. Australian and international scientists met in Perth, Western Australia recently at a Sir Mark Oliphant Conference on the Frontiers of Science, to compare notes and polish their latest high-tech ‘crystal balls'.

There are thresholds in human affairs too – booms and crashes in the money markets, sudden shifts in public opinion, changes in community behaviour, the explosion in the World Wide Web, even the outbreak of wars.

Fascinating, and dead on topic.

Threshold and pattern dynamics uses mathematics and computers to build a model of the factors driving uncommon and important events, Sivapalan explains. "By running these models forward in time, it becomes possible to predict when vital thresholds will be crossed – when things will shift dramatically from their present state to another, possibly dangerous or unstable one.

"For example, if you think about rain falling on the soil, it reaches a point where the soil has absorbed as much water as it can, a threshold is crossed, and the water begins to run off and cause flooding," he says. "Another example of a threshold is erosion, when the power of the wind or floodwater reaches a point where it can dislodge soil particles and sweep them away. A third case is when an apparently stable environment like farmland is hit by rising saline groundwater – and everything suddenly dies."

What we really have to think of right now is this: do we want to build a model on the causes or on the effect? The question is whether we want an endogenous or exogenous system as katz calls it. Technical analysis is all about the effect. We look at the chart and the chart alone is enough. This guy in the quote is referring more to the causes. We don't look at the past flooding to know the future flooding: we look at the causes and how much water the soil can absorb. Well, it might not make much difference. The point is how much it rained. I would not worry about the soil. I would look at how much it rained on some mountain where they have the machine measuring the rain, on the instances there was flooding. That's all. Let's keep reading. But this distinction is important. Causes vs effects. We can't go crazy looking for all the causes making prices go up. We should instead learn to predict prices based on their own past behaviour. At least that's all I've been doing and that's all I can do with my limited tools.

Some thresholds are reversible, others are not – or are extremely hard to re-cross, according to Sivapalan. Hence the importance of having good predictive tools. "The build-up to many of these things is extremely hard to observe, perhaps because it takes place somewhere it is not easy to take measurements. Then the challenge is to identify telltale things we can observe, and see if we can identify patterns in them which point to a future threshold being crossed."
Well, there you go! He says exactly what I said: it's hard to measure all the causes. Let's focus on the "telltale things we can observe, and see if we can identify patterns in them". That's technical analysis.

The task, which scientists around the world are working on, is immensely complex and challenging, but in fields like earthquake prediction, there are encouraging signs of progress.

The science of threshold and pattern dynamics has its roots in the efforts of scientists over the last 20 years to predict earthquakes. Geologists can measure the build-up of giant stresses along critical rock fractures (or faults) deep in the Earth, and estimate the accumulated energy – but knowing exactly when the rocks will slip and how much energy will be released has proven a huge challenge.
Fascinating. Earthquakes. The difference is that I am not really to predict market crashes, which would be equivalent to earthquakes. I am just happy with little changes: i don't need that much volatility you see. I have futures for that - they multiply the moves tenfold.

A team led by John Rundle at the University of California in Davis has developed computer-based methods that forecast earthquakes with much greater precision. "Most people would say that earthquakes can't be predicted or forecast and, indeed, there have been many notable failures," he says. However, his team has overcome the main obstacle to prediction: time. Humans live for a few decades, but large quakes recur over hundreds of years – outside our ability to accurately observe and remember.

By using a computer to simulate the whole fault system, it is possible to see thousands of years of geological activity, Rundle explains. His program simulates the tectonic plates in a fault moving away from each other at a constant rate. After a century or so a threshold is reached where the stresses on the fault are so great that the rocks slip - and a quake occurs. This temporarily lowers the stress on the fault, and the process starts again.

The team still cannot predict the precise time of an upcoming earthquake, "but we can now say that it is likely to happen in one of a small number of areas within a certain time window," says Rundle.

Nice.

Research into thresholds has many possible uses, he adds. "For example: why do countries go to war? Take the United States where, in 1941, it wasn't really until Pearl Harbour that some sort of a social threshold was reached when people en masse decided: ‘We're not going to stand for this. We are going to war'."

A similar pattern occurred with the World Wide Web, which didn't achieve widespread use until a certain level of connectivity was reached. "It looks more and more like this was a sudden process, a threshold. You had to reach some critical level in connections between the computers of the world before the usefulness of the Web became apparent to most people," says Rundle.

Yeah, that's right. I remember in 1997 I thought the web wasn't that big a deal. Now I am almost making money through it. Maybe one day it will support me.

Michael Raupach, an atmospheric scientist with Australia's research agency, the CSIRO, is analysing past abrupt climate changes - from ice ages to salinity - to try and identify the external forces that might cause our present climate patterns and ecosystems to collapse.
There he goes with the "external forces" again. I can't do this with trading. Screw fundamental analysis. I can't backtest it. Too complex.
"With salinity, for a long time while the saline groundwater is rising, you see nothing," he says. "But when the salty water reaches the surface or root zone of plants and trees – the threshold – you see sudden death across a wide area. This is due to a relatively subtle shift in the level of the groundwater."

Another example is the drought which has lately afflicted eastern Australia. The subtle difference from past dry periods was the interaction between drought and warming. While this drought was similar to past events in lack of rainfall, a new feature this time was heat: it was by far the hottest drought on record, because of global warming. This combination pushed many parts of the landscape, including deep-rooted trees, beyond the threshold of no return.

By identifying the external forces that drive such events, it may be possible to predict critical changes and either prevent them or else manage the consequences, says Raupach. His research uses well-understood systems - like fires and stockmarkets - and analyses them to understand the drivers. It has revealed hallmarks common to other complex systems, an indication that there may be universal factors that can be used to analyse all systems.

Raupach's research asks four questions: Are there thresholds? How will things be different if we cross the threshold? What drives the threshold? Can we manage the system to lower the possibility of bad outcomes?

"One of the fundamental questions is: 'Can we identify what the crucial interactions are so we can get some idea of how likely this is?'," he explains. "We might be able to figure out how close we are to the threshold and determine the probability of tipping our climate into a different state."

His research opens the way to better manage human actions that do the greatest damage to the environment and natural ecosystems. "This will mean we are not completely powerless in influencing the trajectory of man-made climate change," says Raupach.

According to Ian Prosser of the CSIRO, drought intensifies the impact of flooding on Australia's water and land-based ecosystems. Heavy floods cause severe erosion of the land and degrade water quality, which in turn can cause havoc in both freshwater and marine environments.

I think they're getting themselves into a mess. But this is inevitable. Because they want to avoid the problems. Unlike me, they can't just make money from going long and short on the natural disasters. That's why we have different concerns. I just care to know if we go up or down. They want to make sure we do not go down: so they have to work much harder than me, to find the causes that make us go down. When instead i am totally ok with going down.

For any future reading and comparison, I need to keep these two things in mind:
1) I want to speculate on future outcomes, whereas these guys want to avoid certain outcomes
2) I want to predict just based on one individual behaviour (earthquake happening or not), whereas they want to focus on all the causes.

For example, I'd wait for an earthquake to begin, and then i'd bet on the fact that it would continue at least for a couple of days, based on past observations. These guys want to know the causes because they want to tell people there will be an earthquake ahead of time. I just want to invest on the earthquake once it's begun.
 
Last edited:
The Secret Life of Chaos


http://en.wikipedia.org/wiki/Turing

Very related to genetic algorithms and trading.

In fact this is related to what I was discussing above. Before we start predicting what will happen in the future, we have to try and code some of what happens with a mathematical formula, which is what the video above is largely about.

1) observe the past
2) code past behaviours
3) test on the out-sample if the code describes/predicts future behaviours

In other words, to paraphrase the title above, we want to find rules and order into an apparent chaos, the rules according to which the markets behave, just like Turing was trying to do with nature. So the video my cousin just sent me via email (he's a physics major and I was telling him about genetic algorithms) is very much on topic, both trading and genetic algorithms.

I also have to investigate this area, to get ideas for my systems:
http://en.wikipedia.org/wiki/Mathematical_biology
Mathematical biology aims at the mathematical representation, treatment and modeling of biological processes, using a variety of applied mathematical techniques and tools.

I am totally illiterate in terms of formulas, but "mathematical biology for dummies" or simiar articles could give me some good ideas.
 
Last edited:
probability interpretations and predictive inference

continuing reasoning from previous post:
http://www.trade2win.com/boards/trading-journals/85510-my-journal-2-a-178.html#post1330742

Now I will read the good old wikipedia:
http://en.wikipedia.org/wiki/Prediction
A prediction or forecast is a statement about the way things will happen in the future, often but not always based on experience or knowledge. While there is much overlap between prediction and forecast, a prediction may be a statement that some outcome is expected, while a forecast may cover a range of possible outcomes.

The typical pseudo-rational trader, who says we're not in the business of making predictions really pisses me off because he is illogical. We are always predicting, any time we're investing. Even if you tell me that you're doing trend following and therefore you're following what the market is telling you, you are still predicting that the market will continue to go where it's been going, so you are still predicting. You goddamn ****er. I always meet some dick head who tells me he doesn't make predictions. Of course he does. He is predicting that his trade will make money, that most of his trades will make money, that his trading will be profitable, or else he wouldn't be trading. There's no way around it. Any trader saying he's not making predictions is a dick head.

Perfect. I kept reading that wikipedia entry and it didn't let me down. Pretty soon it lead me to what i was looking for:
http://en.wikipedia.org/wiki/Predictive_inference
Predictive inference is an interpretation of probability that emphasizes the prediction of future observations based on past observations.
This is precisely the field of trading systems.

Here's a bunch of other related links:
http://en.wikipedia.org/wiki/Category:Statistical_inference
http://en.wikipedia.org/wiki/Category:Statistical_forecasting
http://en.wikipedia.org/wiki/Futures_techniques

And these two major links:
http://en.wikipedia.org/wiki/Probability_interpretations
The word probability has been used in a variety of ways since it was first coined in relation to games of chance. Does probability measure the real, physical tendency of something to occur, or is it just a measure of how strongly one believes it will occur? In answering such questions, we interpret the probability values of probability theory.

http://en.wikipedia.org/wiki/Probability_theory
Probability theory is the branch of mathematics concerned with analysis of random phenomena.
Well, here we have a problem because the markets are not random.

But why am I reading all this stuff? Maybe I am just wasting time. Yeah. I am better off focusing on the threads I listed in the previous post.

Anyway, to wrap it up, here's what we've established here. Mainly what we are doing and its connection to probability theory, through wikipedia links themselves:

Predictive inference < Probability interpretations < Probability theory
Probability theory is the branch of mathematics concerned with analysis of random phenomena.[1] The central objects of probability theory are random variables, stochastic processes, and events: mathematical abstractions of non-deterministic events or measured quantities that may either be single occurrences or evolve over time in an apparently random fashion. Although an individual coin toss or the roll of a die is a random event, if repeated many times the sequence of random events will exhibit certain statistical patterns, which can be studied and predicted. Two representative mathematical results describing such patterns are the law of large numbers and the central limit theorem.

There, it's all wrapped up. Tomorrow I'll focus on these threads:

Gathering ideas from traders
ForexMorningTrade journal
Goldmine!
Alpha Markets - Simple Forex Trading
Spreadbetting FX using TA
My Journal at a Prop house
Entry=Risk Exit=Reward
Vwap Engine

And/or I will focus on asking myself and others: "how do you predict the path this person will walk?" and its application to trading systems.

Whichever one will be easier for me. Probably the second one. Those threads all give me a stomachache. They are the best probably, but still not user-friendly at all.

Manhã tão bonita manhã
De um dia feliz que chegou
O sol no céu surgiu
Em cada cor brilhou
Voltou o sonho então ao coração


Depois deste dia feliz
No sei se outro dia havera
E nossa manhã, tão bela afinal
Manhã de carnaval
 
Last edited:
Re: Getting ideas for new systems

Gathering ideas from non-traders
For this I can start by asking myself the question as follows: if we were predicting the behaviour of a pedestrian, how would we go about calling the direction where he will go? Now, we cannot predict some direction that he's likely to take, because in the markets we have to predict something that is unlikely or else we cannot cover commissions. To do this I would have to observe pedestrians for a while. I should sit at a bar, identify someone walking down the street, and try to predict whether he will turn one way or the other. Then I should look at what reasoning is going on in my mind and if I can apply any of that reasoning to the markets. In the same way, I should ask the same type of question to others. Another good general question is "how well can the past predict the future?". In nature, this is too easy: in winter it's cold and in summer it's hot. I need to study a subject where the laws are not so certain and clear. Something like the markets. I have to make a lot of comparisons with the markets.

Interesting angle. Rather than studying single individuals though, perhaps you'd be able to gain more from analysing crowds and their behaviour? As that's what the markets are in essence...
 
Yeah, too bad my father did not help out, as usual. I was asking him for ideas last night. He was pretty quiet as usual. He used to be a professor of sociology and obviously he knows a lot about statistics. This is totally on topic. I will investigate this field on my own, as you suggest (once I get home):
http://en.wikipedia.org/wiki/Sociology
http://en.wikipedia.org/wiki/Subfields_of_sociology
http://en.wikipedia.org/wiki/Collective_behavior
http://en.wikipedia.org/wiki/Mathematical_sociology
http://en.wikipedia.org/wiki/Sociology_of_markets
http://en.wikipedia.org/wiki/Behavioral_economics
http://en.wikipedia.org/wiki/Behavioral_economics#Behavioral_finance
 
Last edited:
Re: probability interpretations and predictive inference

The typical pseudo-rational trader, who says we're not in the business of making predictions really pisses me off because he is illogical. We are always predicting, any time we're investing. Even if you tell me that you're doing trend following and therefore you're following what the market is telling you, you are still predicting that the market will continue to go where it's been going, so you are still predicting. You goddamn ****er. I always meet some dick head who tells me he doesn't make predictions. Of course he does. He is predicting that his trade will make money, that most of his trades will make money, that his trading will be profitable, or else he wouldn't be trading. There's no way around it. Any trader saying he's not making predictions is a dick head.

That's so true. I never thought of it like that. It's just like the people who say "don't judge me". Of course I'm going to judge them. I'm human, we judge everything.


ForexMorningTrade is hot. Real hot. It's so hot you could fry an egg on it. Which leads to the big question: when and how is it going to fail? Because it will, at some time. It would be a great exercise in observing system failure, to learn how to spot it happening. The question is, how much leverage to trade it with. My instinct says just the usual 100,000 or 50,000 if my equity drops down to $25K - but the little demon on my shoulder says 'optimal f'.

I haven't looked at those other threads. Word has it the Jahdave's journal is also good, and ProSwingTrading - I also haven't looked at them.
 
Thanks for the advice. I will check out Jahdave's journal. Oh, wait. I just went to his journal and you are probably referring to this:
Elliott Wave EUR/USD

I can't get into Elliott Wave. First of all, it's hard to back-test. Second of all, I don't believe it works, and I can't find out because I am unable to back-test it.

If you have any precise information on how the Forex Morning system works, please let me know, here or privately. If I could code it very quickly it would be ideal, and i'd simply add it to the dozens I already have and move on to the next system. If you say it's hot, that means you already know how it works, so please tell me about it.
 
Last edited:
Behavioral finance and predictive inference

This is it, so far the most enlightening wikipedia entry on what I am doing:
http://en.wikipedia.org/wiki/Behavioral_economics#Behavioral_finance

I am going to read this, once I get home.

This is gold:
The central issue in behavioral finance is explaining why market participants make systematic errors. Such errors affect prices and returns, creating market inefficiencies. It also investigates how other participants arbitrage such market inefficiencies.

Behavioral finance highlights inefficiencies such as under- or over-reactions to information as causes of market trends and in extreme cases of bubbles and crashes). Such reactions have been attributed to limited investor attention, overconfidence, overoptimism, mimicry (herding instinct) and noise trading. Technical analysts consider behavioral economics' academic cousin, behavioral finance, to be the theoretical basis for technical analysis...

It is exactly what we are doing, mixed with this other entry:
http://en.wikipedia.org/wiki/Predictive_inference
Predictive inference is an interpretation of probability that emphasizes the prediction of future observations based on past observations.

But the latter entry is short, so let's get back to the first entry:
http://en.wikipedia.org/wiki/Behavioral_economics#Behavioral_finance

Golden entry:
Other key observations include the asymmetry between decisions to acquire or keep resources, known as the "bird in the bush" paradox, and loss aversion, the unwillingness to let go of a valued possession. Loss aversion appears to manifest itself in investor behavior as a reluctance to sell shares or other equity, if doing so would result in a nominal loss.[19] It may also help explain why housing prices rarely/slowly decline to market clearing levels during periods of low demand.

Benartzi and Thaler (1995), applying a version of prospect theory, claim to have solved the equity premium puzzle, something conventional finance models have been unable to do so far.[20] Experimental finance applies the experimental method, e.g. creating an artificial market by some kind of simulation software to study people's decision-making process and behavior in financial markets.

"Bird in the bush"?
http://www.spiritus-temporis.com/bird-in-the-bush/

"A bird in the hand is worth two in the bush."

A proverb of which one meaning is it is better to have something for sure (low risk) than something speculative (high risk). The paradox part is that investors will often not do a probability analysis for lowest risk/highest return. They will take their chances of getting something more speculatively than they will getting something less as a sure thing.

It is well known that investors are risk-averse: Given the opportunity to buy a new investment which has a 50% chance of returning twice the return on an existing investment the average investor will always keep the existing investment (in the example, each investment has the same average rate of return). However, investors will even keep "safe" investments when a riskier investment would on average return a greater profit.

At the same time, people exhibit a herd mentality: If everyone else is buying a stock, then it must be a good buy. If everyone else is selling a stock it must be a bad stock. Contrarian investment strategies argue, correctly, that -- all other things being equal (a major caveat!) that one should buy when all others are selling an sell when all others are buying.

The combination of these two facts -- the aversity of the average investor to risk and the tendency to assume that the future will be like the past -- explains why most investors do not profit as much as they could from stocks and would really be better off investing in mutual funds a relatively safe investment with a better rate of return than bonds (one of the safest of investments ). Risk and potential reward are correlated (risky investments to attract capital must promise better return; safe investments are not under this pressure). The proverb reflects this -- and also the aversity to risk that most humans display. The two birds are of course objectively worth almost twice as much as the bird one has -- if you can catch them that is.

The proverb is a warning to the imprudent and thus reinforces the irrational risk-averse tendency of most persons.

Due to the phrasing of the sentence (and the emphasis given to the position of the bird) it is plain that it is 'not a paradox' -- there is no counterfactual situation or logical impossibility created in this proverb.
****, this is a good explanation, but even too good and I didn't get the goddamn point.

Let's move ahead.

Getting closer and closer...

Models
Some financial models used in money management and asset valuation incorporate behavioral finance parameters, for example:

Thaler's model of price reactions to information, with two phases, underreaction-adjustment-overreaction, creating a price trend
One characteristic of overreaction is that average returns following announcements of good news is lower than following bad news. In other words, overreaction occurs if the market reacts too strongly or for too long to news, thus requiring adjustment in the opposite direction. As a result, outperforming assets in one period are likely to underperform in the following period.
Good stuff, but nothing to create a system on. Or maybe we could look for an edge right after the news release. But here there's a danger of overoptimization. I sense it's not a good field to investigate and I've tested the 14.30 CET time zone extensively, so let's move on.

http://en.wikipedia.org/wiki/Behavioral_economics#Quantitative
Quantitative
Quantitative behavioral finance uses mathematical and statistical methodology to understand behavioral biases. Leading contributors include Gunduz Caginalp (Editor of the Journal of Behavioral Finance from 2001–2004) and collaborators including 2002 Nobelist Vernon Smith, David Porter, Don Balenovich,[23] Vladimira Ilieva and Ahmet Duran[24] and Ray Sturm.[25]

The research can be grouped into the following areas:

1.Empirical studies that demonstrate significant deviations from classical theories
2.Modeling using the concepts of behavioral effects together with the non-classical assumption of the finiteness of assets
3.Forecasting based on these methods
4.Testing models against experimental asset markets
Yes, this is it. My thing is quantitative behavioral finance and predictive inference. What did I get from this immersion in academia? Probably nothing. Just a waste of time. Luckily I don't fall for this bull****. I can tell these academics are good at wasting each other's time with their bull****. Universities are a bull**** factory.

Now I just have to move on towards my objective and task head on, without adopting any side approaches. I have to look on the web for traders and their edges.
 
Last edited:
Yes that's the one. I actually like Elliot Wave because it's got a sort fractal dimension to it, but I have never managed to code anything around it. It doesn't lend itself to being programmed, but then I thought that could be another reason to try.

re ForexMorningTrade, there are enough posts in the thread for anyone to work out what the system does, and in fact it's not even a unique system - the system seller has just optimised it and packaged it nicely on the right trading platform. If you had the right keywords, you could find a lot of info about the phenomenon it exploits on the net, I think. Haven't had a chance to do so myself though. Plus of course that thread is insanely long so I wouldn't recommend reading the whole thing unless you find it interesting seeing how the participants are more like a fan club than a group of traders.
 
Status
Not open for further replies.
Top