Here you may see an enlargement of the chart that the subscribers have receive one week ago, last Friday night (CET), with overimposed the real weekly action: r.Virgeel, in its weekly version, has nicely caught this week range and direction, shown in the rightmost bar, confirming the turn signal generated by the daily forecast.
In circumstances like this, the power of artificial intelligence shows at its best, being able to provide an information that no technical analysis can even imagine.
Has r.Virgeel access to reserved infos? No. The model is built with the numbers publicly available about the world markets and no mind reading ability has been injected in it. 😉
It is just the power of correlation and pattern recognition: if the model is correctly setup, it sees things that we humans… It’s not magic. It’s number crunching.
Following the requests coming from some users that were initially confused about how to interpret the signals coming out from r.Virgeel, I think that nothing is better than to explain how the indicators are produced. I have never done it before so extensively.
The process is very different from the calculations for a technical indicator as RSI or MACD, which involves the direct extraction of values from the price historical data.
Using neural networks follows a different logic: what to I want to know? Think to a medical sample: doctor, I have these symptoms, which is the disease? Ideally, if I have a database of symptoms associated to various diseases, querying the database can give an easily probable range of possibilities, if not the correct reply. This happens now in many major hospitals around the world.
If you have ever used a spreadsheet (who doesn’t?), think to a table made of columns of homogeneous data, each line a single “experience”. You may add one or more columns and associate to each experience a characteristic, a category, a value, a disease, something that give sense to the line of data. Great, you have created an a.i. model. It is a table, where some columns are input data that represent an event or an experience, and other columns that represent a meaning that we want to associate to each event or experience.
Usually, by habit, input columns are placed left in the spreadsheet and associated values are placed right.
r.Virgeel left columns does NOT include any data from the S&P 500 index. All the inputs come from dozens of other market indices, that covers all main stock markets, commodities, forex, metals, yields, bonds, etc.
I may say that r.Virgeel evaluates the S&P 500 as a consequence of all other markets. The influence of each market is directly related to its dimension, so the bond market and the forex are the most relevant.
At the moment the daily model has 74 indices on the input side and the weekly model has 123 inputs on the input side.
At this point, how the r.Virgeel indicators are built? What is placed in the right side columns?
I will cover the main indicators, the rv.Stop, rv.Target, rv.Position and rv.Signal indicators.
First, the market action is broken in “positions”, either long and short. Here you may see a sample of a long position, from the training program. This code is one of the three legs that support r.Virgeel (the others being the data maintenance and the number cruncher): here is where the “experience” is transferred into the model.
The position is marked by the thick white line.
rv.Stop (red dots) is placed to accommodate the whole movement. This is a traditional stop, a value that, if perforated, does invalidate the position.
rv.Target is calculated from the highest high and from the highest close of the final bar of the position.
Two series of signals are placed at significant bars (let’s call them “arrows”): the rv.Position puts arrows at the first and second bar of the move, to mark the entry points, and at the second-last and at the last bar, to mark the close of the position. Then it marks all the other bars of the position with a “Stay long” marker. The same signals are placed in the rv.Signal indicator, except for the “Stay long” markers, plus other arrows are placed at meaningful bars, to highlight minor corrective movements and new entry points available, as the main position is holding.
These four indicators work in accord almost always. Almost. When they point in different directions, it’s a nice warning that something crucial is happening. Big attention is suggested then. After recent improvements, it seems that indecision/uncertainty usually last one/two bars, then r.Virgeel returns under normal conditions.
In the background, you may see coded the rv.PosColor: bright red at major long entry points, bright red at major short entry points, a gradient in between. It is a confirming tool.
All indicators are calculated separately and r.Virgeel is totally unaware of its own past outputs. Every time it produces an output, let’s call it a forecast or a diagnosis, it evaluates present values on the left side against the archive of experiences it has recorded. It is called PatternRecognition and is produced by applying a reverse logic algorithm. The less improbable result is output.
This is how neural networks work. Now you may understand better when I’ve written “A.I. is different“, it has nothing to do with technical analysis by any means.
One recent improvement has been the introduction of the FastTrack indicator. It is the first indicator that has been suggested directly by r.Virgeel. The model has a strange behaviour, probably depending on the fact that I’m not a mathematician I cannot explain it, that produces exaggerated forecasts under defined conditions. Leveraging on this defect, I verified extensively that the output was a fantastic trend follower: false signals are near to nil, considering that even if when they occur, usually you may return on the correct side with a minimum loss. The FastTrack output is a series of four levels. What matters are the two levels that define the central neutrality area.
Let’s make an example. Today’s D-FT (Daily FastTrack) is the sequence 2837.79, 2797.362 | 2808.145, 2788.749
The ordered sequence is 2837.79, 2808.145, 2797.362, 2788.749
The central neutrality area is defined by the numbers 2808.145, 2797.362
Above 2808.145, we have confirmation of a rising dynamic under development,
Under 2797.362, we should have confirmed of a declining action under development.
The FastTrack is an end of day indicator, it is very sensible and I find it useful placed on the short term intraday real-time chart. Actually, using the r.Virgeel’s framework, the D-FT levels, some rudiments of Elliott’s Wave’s theory and a couple of moving averages seems a good setup. I’m testing it at the moment with a certain satisfaction and I will report the experience when it will be mature. Simply said, I’m moving the first steps towards a wealth building machine, using intraday action, based on a weekly staircase. Each week target is based on a stable compound capitalization model. In a few month I will go bankrupt (on the CFD account) or the machine will be snorting and puffing and chugging.
r.Virgeel’s indicators somehow mimic some traditional t.a. indicators and this may generate confusion or misunderstanding. It’s my fault: as the model construction has developed, I’ve got from my previous experience of investor and trading scholar, fishing ideas trying to shape r.Virgeel. Many indicators have been developed and few survived. Some brilliant ideas have been demonstrated them dumb, others have a long track of solidity and sharpness. The quest is not over.
In the previous post “Forecast/ability” I did refer to the daily a.i. forecasts and I showed the results of a long and extensive research on the quality of the response of the model.
But when we come to the weekly and to the monthly forecast, things change radically and for the best. Undoubtedly, weekly and monthly bars undergo a reduced “noise” and express better the global consent of the partecipants to the market activity. Market is fractal in nature and I have not an explanation for why it behaves differently from daily to weekly and monthly time frames. Maybe because it reflects the attitudes of different categories of actors (investors have a totally different approach to the market than daytraders or position traders). Anyway this is what comes out years of observations of the forecasts produced by my model.
Just as an example, this monthly forecast chart has been produced exactly one year ago, on 5th of December of 2017 and shows how r.Virgeel forecasted correctly the October 2018 correction, ten months in advance. Astonishing, uh? Consider that the monthly model is a long term database of financial and econometric data, so the detected patterns are not only related to the market activity, but also to the underlying economic activity.
(The right part of the chart has been cutted, for respect to paying subscribers, as it refers to current market expectations and it is still valid).
Here another example, from the weekly model, published to subscribers on the 3rd of March 2018: the deep and scary correction was shaking the markets and r.Virgeel correctly forecasted that in 4 to 5 weeks the S&P 500 should reverse (not reaching the previous low) and go for new all times highs, as it did.
Quite obviously, these forecast on the long term side are not much interesting for very short term traders, but they may be invaluable for position traders and investors, that have a global vision of the market totally depurated from the biased news and the so-called experts opinions. hese are not opinions of any kind, they are the result of pure brute force number crunching
The psychological advantage of these knowledge is actually the first and best result: less stress, better decisions, more returns on your investments!
Following my previous post “New Tools at the Horizon“, one question was twirling in my mind: why the stock market is forecastable, but the forecasts are not affordable?
The forecastability of the market is an evidence, because if it were not – being it just a random walk – there would not be the possibility to have an output from the neural networks that manage the forecast process. For a neural network to work, there must be some sort of structure inside tha data that can be used to produce the forecast/diagnosis.
This chart shows a blind neural network, unable to recognize any pattern in the input data.
And this hidden structure is present indeed inside the market data, otherwise r.Virgeel would be totally blind and dumb. This is a sample chart of a blind network: not structure is evaluated and the output is just an array of zero values.
The fact that we humans do not recognize any structure in the data is irrelevant.
So we have a (hidden) structure, the neural tools recognize it, but the output ranges from nicely precise to totally incorrect, without having the possibility to know how much the result is matching the real future movement of the price.
Now, I begin to see the light.
The price of a financial instrument is the result of an ask/bid process, where a multitude of actors (I’m considering liquid markets with a wide audience) buy and sell that instruments under the suggestion of a personal forecast that the price of that instrument will rise or fall in the future. Every partecipant to this activity actually does a personal forecast every time he/she executes an order. So, the resulting price is the sum of all the collective forecasts and, at the end of the day, this collective forecasting process generates the push that contribute to move the trend.
[revec2t text="Every partecipant to the market activity actually does a personal forecast every time he/she executes an order."]
In other words, every attempt to forecast the market is a process of forecasting a collective forecast activity, a meta-forecast: no surprise that somewhere in the process one or more dimensions are lost and the result is probably something similar to a shadow, that let you recognize the original shape under certain conditions and totally mistify the original shape under other conditions. When you project a multidimensional event in a field that reduces the dimensions (think to a 3d object projected onto a plane) you lose a significant portion of information and you may generate a lot of ambuguity.
A 3d object projected onto a 2d planes may generate very different shapes
Now, the forecasting process is just a minor side activity of r.Virgeel, even if it is the most appealing and mind-storming: r.Virgeel is mostly a diagnostic tool that reads current data and find historical patterns that match the best market position available, with a significant success.
I’m fully aware of the basic fact that every trader and investor has it’s own style. Many school of thinking, but everyone is really different, particularly in the private sector. If a private trader survives the first 18 months without being wiped out, then she/he may have the possibility to play on. It’s a sad and real statistics: 95% of traders are wiped out in the first 12-18 months of activity. So, if you are a private trader since years, you are in the 5% and you have built your skills with iron and steel, it’s not easy to approach a new language.
In effect, a.i. tools are very comprehensive: the result is there, fast, precise, direct. You just have to pay attention, when required, then to act if necessary. The whole decision process that makes the foundations of a trading system is questioned. It’s not an easy process, it will not be.
Mind changer about traditional tools as t.a.
Artificial Intelligence tools, not only r.Virgeel, implies that you change your mind about the instruments of analysis. Are you there with any of the technical analysis tools, or with a realtime CFD, you plot your oscillators, you backtest an optimal setting and decide, based on a unique line of data and a monodimensional analysis. A.I. can analyze many different correlations concurrently, sorting out with surprising results. It is obviously different, it is much different.
If you are totally unaware of a.i.. and neural networks, you do not need to read huge books to understand. You use Excel, for sure. Read this, then it will be easier. Made simple, supercharge and megapower a spreadsheet and you have a neural network. Then you have all the a.i.: image recognition, diagnostic, speech recognition, text recognition, and you name it. Once you understand the brick, then the wall makes sense.
The fact is that an a.i. robo-advisor makes the dirty work for you. Collects data, builds tables, generates logic, connects correlations, normalizes the resulting cloud of data and generates the output data. Whooo! All in one. And it’s fast!
Low adrenaline, stress reducer
People love adrenaline, excitement and, basically, confusion. In the long term, this attitude generates deep stress, that reflects into bad behaviours. If you are a few minutes ticker trader, r. Virgeel is not for you, probably. Well, as it gives a daily trend, it may help, but…
Once you have a good robo-advisor, you need less to have opinions about everything. It monitors the data, it makes the loops, it builds the report. You must trust it, follow the suggestions and pull the triggers. Stress is highly reduced. Instead of needing to have an opinion about everything, you just have to evaluate a very well documented “opinion” generated by the a.i. model. You know, I like to say, it’s different.
For what I can see, robo-advisory is taking shape as large (and expensive) premium sites with a large choice of instruments, with applied different forms of analysis and previsions. I argue that the neural analysis is often simplified and this can be a good thing, overall. In other words, these sites offer a large set of instruments and an automated training.
Here at spxbot it’s a completely different music (well, it’s a premium site, but it’s cheap). First, just one instrument – the S&P 500 – here. Then, the correlated instruments are hundreds, to represent the global finance world. All for one, we may say 😉 More, r.Virgeel is highly deparametrized, meaning that it work by brute force and not by rules. Finally, r.Virgeel is trained “personally”, bar by bar. Total human supevision on the training of the model. You know, it’s always the same old story: you put rubbish in, you get rubbish out.
This post explains the main website feature: the indicators that form r.Virgeel vision of the market.
The indicators are:
Bars ahead – neurally calculated – H/L/C is forecasted for the next 24 bars Target– neurally calculated – where the current move is heading Stop – neurally calculated- a value that confirm the trend and generates alerts of reversal Position – neurally calculated – a simple, but detailed, as a neural swing system, Position is calculated in three fashion: the positive attitude, the negative attiture and an overall attitude. In moment of uncertainty this help to have confirmation of the reading. Color Bars – neurally calculated – an evaluation of market’s potential energy through color code. It has integrated the old Stamina indicator. Signal – neurally calculated – top/bottom pattern recognition. It’s the simplest original version of the Position indicator: it fires probable tops and bottoms detection. FastTrack – The standard report integrates now the FastTrack levels, from daily, weekly and monthly models for SPX. The daily D-FT is available for:
S&P 500
Dow Jones Industrial Average
Nasdaq Composite
DAX Frankfurt
SHC Shanghai
GOLD
OIL WTI
All indicators are calculated indipendently from one another, giving the opportunity to let them reciprocally confirm.
The text on the left of the chart summarize the relevant numerical data, more numerical data in the body of the mail/post.
BARS AHEAD
Coloured bars. It’s a glimpse into the future: they represent the less improbable path the market is supposed to follow, for the next 24 bars. In the background, the latest monthly and weekly forecaste bars give a more complete and synchronical view of the coming events.
TARGET
Blue dots, blue lines. It’s a glimpse into the future: the Target is an evaluation of the price level that the S&P is bound. More the price nears the Target, higher the probabilities of an imminent turn.
STOP
Red dot, red lines. It’s a reading of the present. The Stop is the value that, if broken at close, suggests that the position has come to an end. It works either for long or short positions. The Stop is free to fluctuate and, by experience, the descending Stop is a sign of strength of the long position (and reverse), usually occurring during choppy phases.
Also, the weekly Stop is present on the chart (darker red band).
POSITION
It’s a reading of the present. The Position indicator is the most evolute and mimicks a complete trading system, with entry signals, position confirmation and exit signals. It is calculated separately for the long and for the short positions.
The Position indicator provides the following signals:
Open Position – generates signal triangle
Add to Position – generates signal triangle
Stay in Position
Reaching Top/Bottom – generates signal triangle
Close Position – generates signal triangle
that follow the ciclical activity of investing and trading.
SIGNAL
It’s a reading of the present. The Signal indicator is the parent of the Position indicator and it was, at a certain point, removed from the charts and dismissed. It has the aim to detect the market turning points. Lately, I decided to revive it and use it the signal generation: you may see, usually near the extremes of the chart, there are some small colored triangles. These triangles are generated from the Position indicator and from the Signal indicator and they usually coincide. The triangles should mark the extremes, but more often they generate a cloud of signals around the reversals.
Following my previous post, I would like to point out that approaching the a.i. advisory, you have to change your mind. With most probability, you are trained in technical analysis, various techniques to train your eye and numbers to correlate the stream of data. With an extended application to chart reading and some discipline, it’s not hard to avoid largest mistakes and trade with some results.
What happens is that your carefully brewed technique, a certain day, blows up. You where riding the wave and … splash! All technical analysis let you ride one wave, if well set up, but you are bound to splash as soon as the new wave comes in.
Try to experiment with volatility and then you begin to see how impredictable are the consequences of human behavior. And that is just one dimension. So, if you are so expert to radically change your trading attitude and timely, no need to read further.
I have to say, I’m not in the group. I tend to stick to that particular solution and trust and be deluded. And lose money. I do not really trade with t.a. any more, except a minimal account where I experiment and keep updated the relation from the a.i. model output and t.a. charting. There, I trade using the a.i. daily forecast as a long term preview and I use a 2h bars chart with my beloved modified DMI plus Parabolic and Hi-Lo Activator. In other word, these are momentum and stops. I activate them manually, but following indicators notifications. Shaking all information, it’s not hard to catch good opportunities, always in the direction of predominant trend. If in doubt, stay out. The account is growing.
So, if the use of a robo-advisor is in your future, prepare to radically change your trading and investing behaviour. Sit down, try to imagine all the things that you can do in your spare time and start. Yes, you will have a lot of spare time, embarrassing lots. You need to fill that time and with periodicity, every day, or even every week, maybe every 4 hours? (it will depend on the advisory you will adopt, there are many around, they are popping out massively), you will just peep, take note and check and take few, very few triggering decisions. Than, fast back to your gardening, to your fitness training, to your passion with motors, to reading as there is no tomorrow. In my spare time, I develop the spxbot a.i. engine and this website.
Just a final note: anytime I tried to use t.a. indicators (and I tried many) inside the model, the result were disastrous. The one dimensional constraint that they introduce conflicts with network correlation ability. In effect, I’ve earned from my really first experiments in neural networks modelling with BrainMaker down in the late ’80s, neural networks work good with the data as raw as possible. Any transformation introduces a field of possible wrong solutions and divergencies proliferate. The neural networks do not need to be perfect, they need to see just a little bit better than you and to discriminate using details that you cannot even imagine exist. That’s not perfection, that’s sharpness.
If you are a happy trader, you can avoid reading this post. You have your instruments and techniques and take home your living. You are in the 5%.
This post is written for the other 95%. Yes, 95% of traders go broke, in the first 12-18 months of activity. The market is merciless with the fool and his money. Being in large company does not help.
It’s not easy to manage a losing position. Taking position requires a lot of opinions and more your position is on the loss side, more opinions are required, in conflict with the original ones. It soon becomes a mess. The psychological stress then goes wild and compromises not only your account, but also your relationships and your social attitudes.
Many famous book writing traders say that you must plan your trading activity and strictly stick to the plan. This undoubtedly reduces errors, but requires a very rational attitude that not everyone has. Usually, a plan is related to technical analysis (t.a.) indicators that nowadays are largely insufficient to guarantee you from errors. Some are better than others, but market is ever changing and the analysis of a single price line on a single time frame is unable to catch the full complexity of what’s going on. Whenever you back test a t.a. system, you are enclosing in such a number of constraints that soon the system will go broke. Maybe some exception is running out there.
Financial markets are complex, being the sum of a huge number of actors. What sorts out is that market is moved by anticipation of future events, through counter opinions and this phenomena globally unfolds in trends. And you see the trends follow one another. Where there is repetition, often you have patterns in action and pattern are generated by interacting frequencies. As a trader, in front of charts, you soon develop the ability to recognize repetitive graphic patterns and maybe you also have some tools that helps you classify the patterns. Charts are fantastic for a fast global opinion and worth nothing for suggesting the correct intervention. The chart is just a slice in the ham of the market: for as good as it is, you are missing the most.
Now, suppose that you have near a good relaxed trader, one in the 5%, and you do not understand very well how he acts, but you see him opening and closing positions, earning money and reducing losses to ridiculous phisiological amounts. You will just follow his steps, soon stepping aside of you opinions, recovering self esteem, reducing stress. It is now much easier to manage losing positions: sometimes hedging, sometimes holding, sometimes taking the (small) loss.
I have begun to build the robo-advisor long before the word was even coined, in early ’90 with the first experiments and in early 2013 with the actual running model. The model is pure brute force pattern recognition and it is under evaluation, by my subscribers and me, no back testing of any sort, since two full years and it is acting nicely. It is a much better trader than I am, so I seat behind and peep. I see it may occasionally be blind or wrong, sometimes late and sometimes early, but, you know, nothing is perfect in this world. It recovers very fast, it is adaptive, it is unbiased, it is responsive and a bit conservative. It keeps you on the sunny side of the road. Stress is over. I see to the charts and it’s easy to recognize the opportunities, manage the positions and trigger the orders. Since long without a name, I now call my robo-advisor r.Vergeel.
PS. a kind reader has made me note that, out there, the number of websites offering a.i. advisory is exploded: I just want you to be aware about many very superficial analysis. Please, make your due homework and consider that single data line analysis is available through well known software and largely insufficient to represent the complexity of the market.
Trading is like skying. You always make the same simple moves (start, end, turn right, turn left) and the same tracks, but never in the same way. Never the same weather, snow condition, visibility, temperature, and so on: every time it’s different.
Imagine that the neural network knows the track. It does not guarantee that you arrive downhill integer, but it can avoid a lot of errors. Just the track, with slopes and boundaries and gradients all around. It has already descended that track and even some variants and similar. It can guide you. I should call the model r.Virgeel.
Now, in effect r.Virgeel knows also the condition of the snow, the visibility, the temperature and parameters, so it can even tell you that it’s better to stay home and sip a cordiale.
Mr. Market, as any good skier, will never do the same path twice, no matter how well r.Virgeel knows the track, but Mr. Market, as any good skier, will stay inside the track. Different paths on the same track. A matter of patterns, with a lot of variables.
A human can hardly conceive that this is even possible, while the neural networks crunch numbers. It’s a matter of data. Well collected and trained data.
By some extent, we can say that r.Virgeel can see into the future. Actually, he can show how market reacted under conditions similar to current ones. It’s not a superpower, it’s brute force. Something that a computer can do, not a human. So, r.Virgeel can show the bars into the future as the less improbable path that Mr. Market is going to run, inside the current track, under current conditions.
Artificial intelligence is for lazy investors and this for two orders of reasons: it reduces the frequency of trading actions and it relieves you from the necessity of having an opinion.
Any of our action requires an opinion and, in trading, our opinions are, usually, like TNT in the hands of a chimp. It’s a big problem. A lot of opinions to have, on gold, on oil, on the stocks, on the banks, on EVERYTHING!! A Correct Opinion!
I Must Have A Correct Opinion On EVERYTHING! Otherwise I Will Lose Money!
In fact, it ends up that we lose money, confidence, respect, with rising tensions and crisis of any kind around, and still we have a very confused opinion on something, ignoring all the rest. Our frequency of trading rises to syncopated and the crash is near.
Now, what A.I. provides is the possibility to setup an environment that classifies data autonomously, large bunches of data without human intervention. You can say that it “learns”, that it “recognises”, that it “sees”. Sometimes, these pieces of code do something that seems magic, as the possibility of reading and forecasting the bars of the S&P 500 into the future.
spxbot.com does this. Far from being perfect, I manage the art of being approximately correct. Neural networks does a reverse reasoning: they find the result as the less improbable value from the data it has been fed with. So, if you can manage a nicely huge amount of data and you can setup the code environment to accomplish the analysis and you have some years to spend on research, I guarantee that you may get some very interesting results. Or, you may subscribe to spxbot!
The brain has passed more than four years of development and it is now nicely working since a while. Signals have a validation sequence, then acting is suggested. Always, it is suggested the day before, to act on next day market. I record market opening as the execution price, just for having a fixed schema. You may act as you prefer, following your trading habits and indicators and even timeframes.
I consider our portfolio having two components: a part dedicated to long term investments and a part for playing trading. Signals from the a.i. indicators should cover both. I say should, because the “brain” is running since less than two years, so it’s behaviour in a falling market is still to be seen. Long Term speaking, the model is still long at the moment.
Since it’s inception, dating back to 02/12/2016, the DTS has taken 12 closed positions, 9 positive and three negative. The three negative position where closed at -0.28%, -0.84% and -0.04%. The global performance of these twelve positions amounts to +34.52%, compared to the SPX +30.80%, during the same period.
Almost one month ago, this daily forecast was correctly warning of the incoming A-B-C wave. Does your tools do something even similar?
Trading is one of the easiest activities around: few seconds and you are in position, few seconds and you are out of your position. Making it profitable is a totally different discourse. Libraries are plenty of technical and fundamental analysis books and the internet is plenty of websites that offer amazing returns. All of them (except some you can count on one hand’s fingers) do the same one dimensional analysis, based on the tape, the fluctuating price of the chosen instrument. The variety of tools available is confusing, but basically they all work on moving averages and momentum, sometimes volatility. So, you have, basically, one dimensional analysis based on momentum. And you lose money.
Two very common profitable trading techniques are insider trading and front running. Both are illegal, and you must be inside the financial business to be able to put them in practice. And if you are not, are you aimed to be stripped? Probably, yes.
Today, artificial intelligence offers an edge for common investors and traders: opposite to the plethora of tech analysis software, it makes a totally different analysis, it needs years to be developed proficiently, it sees things that the human eyes cannot see.
Here at spxbot, I have taken this totally different approach: the model crunches long term data from dozens of different inputs, and when I say dozens I really mean an approach that is impossible for a human to even imagine. The daily model, that forecasts the daily time frame, has 87 basic inputs, at the moment, that cover almost every aspects of the financial markets: stocks exchanges, commodities, rates, bonds, metals, currencies. The world spins and the data gets in, then the data is prepared for the model, trained, tested, verified… well, it’s a long process, but computers are fast and after years of development the code is well fitted. After data preprocessing, we are talking of a “brain” with millions of neurons at work.
What sorts out is stunning, something that makes momentum trading appear as old as a sumerian cuneiform clay tablet. Artificial Intelligence is able to recognize real patterns inside the market hidden (and evident) disorder and give back an unbiased, unemotional reading of the future events. Believe it or not, A.I. can snoop incoming events, in advance. The spxbot model is designed to be adaptive and responsive, and built to accomplish its tasks without any human opinion intervention.
It’s not my aim to explain how Artificial Intelligence works and I never tried to, just sometimes I’ve discussed the media approach on the topic.
I’ve seen this video and it explains quite well how the core of this technology works. We are speaking about back-propagation neural network, the pattern recognition engine that works at the core of r.Virgeel and that makes the magic.
My first experiments with neural networks were back in the early 90s using Brainmaker. It was not 4 minutes, but in maybe half an our you could have a quite complex network running. Then training time was hours. In 2013 it took me three years to build the spxbot.com model and it’s under continuous development. Large part of the work really stands in maintaining the data base, formed by tables that have hundreds of fields and thousands of records that must be aligned and updated, always. Without exception. Now the network size has reached the remarkable dimension of a maximum over twenty millions of neurons. It is not the positronic brain, yet. It is just a single complex table that offers the possibility to correlate a large number of inputs and some experimental training ideas for the output side.
Do your remember Visicalc? Probably not. Visicalc was the first spreadsheet, 20 rows by 5 column (!) by Dan Bricklin with Bob Frankston. It was 1978. The Spreadsheet! The tools that makes the world go around… You have a world before 1978 and a world after 1978. And the world after 1978 has transposed in the spreadsheet programs (now thousands of rows by thousands of columns, and in the cloud!) what they were used to do by hand.
Neural networks (the logical and computational engines of AI systems) are not so different from a spreadsheet. In effect, they are spreadsheets! And as spreadsheets, they get in raw data to output a result. Simply, neural networks contains the data and a relational tool that correlates the data, and this engine (the dark box) apply a reverse logic. Visical and all its children and nephews apply algebra to raw data and calculate the result. The neural networks do not calculate the exact result, they calculate the less improbable result, in the range of the previously given results. The two, often, coincide. It is, anyway, a statistic tool.
The logic of “less improbable” has demonstrated to be very effective in solving very complex problems, like image analysis, categorization of large amount of data, correlation of many different inputs. Neural networks are a powerful diagnostic tool: given symptoms, a range of possible deseases is restricted; or think to recognizing elements in an image, here they do real the magic: the neural networks can see. They can recognize fingerprints, faces, animals …
In the case of trading, the neural networks can provide an unbiased unique reading of the market.
I’m wandering: do we need Artificial Intelligence to recognize that we are in a bull market? When you are inside a trend, corrections are our friends and the fear of the top (or bottom) is our recurrent trouble. But, surfing is easy with a trendy market. Stay hungry, stay long.
I imagine that many readers interested in the Artificial Intelligence topic are getting used to a very romantic view of the argument. The majority of the articles you may read present anxious questions about a technocratic future generated and managed by AI driven machines. Is this wrong? Not at all, but this is the romantic side of the story. Many of us will surely loose their jobs due to “intelligent” machines, quite soon. The production/consumption paradigm is rapidly changing, as society adopts growing levels of robotics.
What really is AI? We are in the field of software programming. Since the beginning, programmers were faced with the problem of transferring the execution of tasks from the material to the digital world. Many tasks could be translated much or more easily, but some tasks could not be solved with the tools that were developed in the first decades of computer science. Then a new concept appeared: neural networks. The current logic is reversed inside a neural network: instead of calculating the result from a procedural sequence, replicating the assembly line in digital form, the result is calculated as the less improbable from the library of possible results.
So, to have a working neural network (an AI system is usually made up with many, that accomplish different typologies of tasks), you must have a database (the larger the better) that collects the data that represent the events you are trying to analyze, plus a fixed algorithm that interpolates the result from the data contained in the database.
This approach to problem solving based on the less improbable solution has demonstrated to work very efficiently on otherwise impossible task, as extracting information from images, ranking and categorizing information and also forecasting complex-ever-changing phenomena, such as weather or the stock market.
What should be clear is that neural networks and all the consequent AI development are basically a statistic tool, a different way to smooth and integrate large quantities of data. The fact that these tools can work so well in apparently impossible tasks is the key of the door that opens on the world of magic. In effect, neural networks seems magic because they can correlate a quantity of data that we cannot. It’s brute force statistic. In you put garbage into a neural network, the output will be garbage. If you put in well tempered data, you may get a surprisingly magic output. It can “recognize”, it can “view” and it can “forecast”. But it remains brute force statistic.
Often, you read someone that has the key of the “hidden order” of the markets and is so kind and altruist to share such a knowledge with you. Can you trust? Can you believe?
In my opinion, after years of testing and developing the artificial intelligence model that powers this site, there is no hidden order of any kind around the market, but, on the other side, you do not need any (hidden or explicit) order to profit from markets. You need to be logical, to reduce risk by knowledge and to act with proper timing and then profits will come.
The inner disorder of the market is generated by the interaction of thousands and thousands of humans concurrently acting on the market, moved by many different passions, from different locations, with different habits and all with just one idea in mind: profit. Trying to reduce this mechanics to the oscillating move from fear to greed and back is simplistic, as many passions act at once in each one of the participants to the market.
This chaotic and never ending activity may be analyzed in terms of correlations and this can achieve great results: simple correlations can be managed by single humans, but complex correlations cannot, only computers can put everything together and try to follow the path of correlations. And correlations change in time, as everything else, and so even computers have hard times to complete such a task. Are rising rates bad for stock market? Maybe. And maybe not. So many factors contribute to our opinions, that our opinions are always weak. The great and, at the moment, irreplaceable feature of computers is that they remove passions (hidden or evident) from the analysis and they can correlate so many different inputs that no human can even imagine to do such a thing.
I read an article today that highlights the fact that a lot of investors are scared by this current stock market. The stock market is moving opposite to almost any prevision made by mainstream media, it should have crashed after Trump election or Brexit, and instead it is rising madly and without rest. All this rises confusion and uncertainty in private investors, which have no support for their investment strategies, except some old and outdated models.
The new frontier of investment behavior is Artificial Intelligence: abstract and unbiased models that can analyze the market without emotional involvement, that can weight huge past historical experiences and that can do correlations no human can even imagine. This is no sci-fi, this is NOW! Forget news, suggestions, best performers or “insider’s tips” of any kind, that just makes you waste money.
Having an A.I. consultant gets you free and makes it easy to recognize how market is moving and where it is going. Here at spxbot.com I analyze only the S&P 500 index, just that, and correlating dozens of financial instruments I can offer an unique and totally unbiased reading of the market and related expectations. And it’s cheap, too! The stock market is running wild and going far, don’t miss this opportunity.
In the rising market, you are continuously convinced that a crash is imminent and this is the fuel that makes the market grow, having a lot of people on the wrong side. We are watching it in the latest months, whenever the downturn seems inevitable, the market turns up blindly, no way.
If I should rely on my own analysis, I would be a net looser. Luckily, the model seems working nicely, even if with some indecisions. In latest days it has flipped from long position to alerts of an imminent turn, but it has never lost the positive attitude, it was just moving on the edge, and you have to consider that the market is in a never seen before territory.
This confirms the ability of the Artificial Intelligence models to manage never seen before events, sometimes better and sometimes with a bit of indecision, but always keeping the correct orientation and pointing towards profit, in our case. Not bad, not bad at all.
Where is the target of the move, now? It’s early to say, probably. Short term, we may consider the area 2330/2360 the place where something might happen, and it can even extend to 2400. The A.I. model is adaptive, so it is not bound to any preconception: everyday it analyzes the numbers and gives its view and I am not able to preview its behavior. I just act after its advice.
Long term investors may sleep quietly, the time for the big turn is yet to come.
Every trader and investor has her/his own framework of opinions that pilot the triggers. It is a mix of experience, knowledge and emotions. As your decisions are worth a gain, you have an euphorical sensation that you are mastering the forces that drive the market, but on the other side if your decisions take you on the losing side, you are bound toward depressions and your trading decisions gets worser and worser.
Every trader or investor has passed through these two extremes of the maniac-depressive syndrome and it is crucially importante to learn to manage its effects.
First, never blame someone else for your faults. Try to recognize your errors and learn to manage your money with realistic expectations. We need to study, not just install a charting software. But there is always a weak point in the process and it is our opinions. Opinions are strictly tied with sentiments and none of us is able to separate the actual objective facts from our perception of them, and this is inevitable.
As traders or investors, we need tools that help us stay away from emotions and from opinions. It may seem strange, but the less opinions you have, the better trading decisions you take. I have learned it developing the artificial intelligence model that powers the forecasts: you do not need to have opinions to make good trading decisions – more, it’s better you do not have opinions at all. No opinions, no stress. No stress, better trading decisions.
Technical analysis is based on opinions, Elliott’s wave theory is based on opinions, Gann’s theory is based on opinions, even fundamental analysis is based on opinions: all these instruments takes your emotions out of control, sooner or later.
Technical analysis, in its various forms, can be very useful: it may provide you with valuable information and can trigger signals in quite an objective way, but only if you use it inside a wider framework that provides you with a deeper reading of the market. Otherwise, it just will take your emotional side out of control.
Having a tool that shows how the market will act in near future is a powerful opinion deleter. No tool is perfect, of course, but if you trade with an unbiased vision of the future, your decisions are free from emotions and your rational expectation are stress free. This is the real and objective power that rises from artificial intelligence applied to trading or investing.
It is almost one year that the spxbot.com site is on and three and a half years since I begun developing the model. Well, many previous experiences have built the necessary skills, but this is the timing. No, it’s not time to draw conclusions, the work is still going on – will it ever last? – as I’m now developing a new weekly model, pretty different from the one that is active at the moment, so to have not one, but two readings and a better vision of the market.
Snooping around I see a lot of bearish attitude and I remember that about four years ago I was very bearish and the market was continuously rising. My incapacity to be well tuned was obvious. The results of the model did surprise me: I have to be sincere, I wouldn’t have bet a dime on its success – market is unpredictable is the mantra we have grown with and it is difficult to reset the way we think.
Technical analysis is almost unuseful, otherwise it would be plenty of millionaires around. Maybe you know, 95% of market private traders lose the capital and the shirt in 12/18 months and they usually trust technical analysis tools. I’ve been subscribed to “T. A. of Stocks and Commodities” for years and it gives you a lot of ideas and strength and good attitude, but then I realized that digital technical analysis is about half a century old and it must have been outmoded. Really do you think that with a line of data and a bunch of indicators you can get rich?
We need instruments to manage market noise and dig inside complex ever changing relations, as society itself is deeply changing, now faster than ever. We need an indipendent view on the market, abstract and unbiased, opposite of the continuously manipulated information we are subject.
PS: Years ago, I used to think that market is where opinions are transformed in profit (or more often, in losses). Now I do not let my opinions hinder my investments, as the model is so much smarter than me.
Just a reminder: about one month ago I made public some of the projections of the A.I. model in the post https://spxbot.com/2016/10/20/euro-investors-may-get-double-profit/ The post was not about the numbers, but about an unique opportunity that Euro based investors were facing (and are still facing).
In the meantime the EURUSD has passed from around 1.095 to 1.06 and the SPX from 2140 to 2190: they sum to a nice 5.5/6 % in just one month.
I’m always here to show something obscure coming in from the future, but today I’d like to to just fix a point in the present.
No divinations. The chart is speaking. This is a CFD continuous chart daily bars of S&P 500 index. We know the market is discounting a crucial event in less the a couple of weeks. The coming of the event has repriced the index, and we are NOW, Friday morning 28th of October, at the equilibrium point. We know one sure thing, that this is the price of the pre-us election expectations of the markets. To make it simple, CFD at 2130 and SPX at 2136. We are having rising noise around this value. One week more.
PS. Either wins HRC or wins DT – In first case, the big business will continue as usual . In second case, the promised fiscal and autarchic reforms may induce a boost in US richness. So, from investors point of view, is this a win-win ?
On 23rd of September the model has produced this forecast. It was Friday, and usually as the week closes the model is sharper in viewing into the future. As well as it seems that Thursday forecasts are the weaker, and here I just guess an interpretation: Thursday reflects all the noise accumulated during the week, while Friday is the day the market moves in “safe” mode for the two days close ahead. Moving to safe mode reflects the real sentiment, unavoidably.
If you check the prevision against what has been the actual market behavior, well… I leave the conclusion to you!
Here you may see an enlargement of the chart that the subscribers have receive one week ago, last Friday night (CET), with overimposed the real weekly action: r.Virgeel, in its weekly version, has nicely caught this week range and direction, shown in the rightmost bar, confirming the turn signal generated by the daily forecast.
Just as an example, this monthly forecast chart has been produced exactly one year ago, on 5th of December of 2017 and shows how r.Virgeel forecasted correctly the October 2018 correction, ten months in advance. Astonishing, uh? Consider that the monthly model is a long term database of financial and econometric data, so the detected patterns are not only related to the market activity, but also to the underlying economic activity.
Here another example, from the weekly model, published to subscribers on the 3rd of March 2018: the deep and scary correction was shaking the markets and r.Virgeel correctly forecasted that in 4 to 5 weeks the S&P 500 should reverse (not reaching the previous low) and go for new all times highs, as it did.
No divinations. The chart is speaking. This is a CFD continuous chart daily bars of S&P 500 index. We know the market is discounting a crucial event in less the a couple of weeks. The coming of the event has repriced the index, and we are NOW, Friday morning 28th of October, at the equilibrium point. We know one sure thing, that this is the price of the pre-us election expectations of the markets. To make it simple, CFD at 2130 and SPX at 2136. We are having rising noise around this value. One week more.
On 23rd of September the model has produced this forecast. It was Friday, and usually as the week closes the model is sharper in viewing into the future. As well as it seems that Thursday forecasts are the weaker, and here I just guess an interpretation: Thursday reflects all the noise accumulated during the week, while Friday is the day the market moves in “safe” mode for the two days close ahead. Moving to safe mode reflects the real sentiment, unavoidably.