Technical Analysis alone will not give you an edge

Good evening my fellow traders, I have been a student of the market since 2011 when I joined FF. The entire concept of investing, and obtaining profit through fluctuations in price movement has fascinated me. I have not been fortunate enough to blow accounts as many of you have, I enjoyed relative success in the markets earlier on which fueled my hubris. I erroneously surmised that observing charts, learning patterns, using some sort of money management system and discipline was responsible for my success. I have studied Technical Analysis tools extensively, from MACD,Stochastics, Parabolic SAR, RSI, S/D, S/R, candle stick patterns, you name it.

I have began taking trading seriously lately, not just as an interest and a passion but to dedicate my time in this as a second carrier. I see many here purport to be able to predict price, to "follow" the market, many here will tell me that you have learned the skill to do this successfully. I will respectfully disagree with you, I go as far as to call it Bull shit. In any field of human history where a phenomenon is not well known, not understood, or plainly random... there will be superstition, there will be an array of different methods, some conflicting that will only work due to pure chance. I claim yours is no different.

When studying any such phenomenon, the best tools at our disposal is the scientific method... 1- ask a question, 2- do research, 3- formulate a hypothesis, 4- test your hypothesis, 5- evaluate your results, 6- communicate your findings.

From readying probability theory and statistics I claim that one cannot profit from random data, do you agree with my claim? if you do not let's discuss your ideas here as I am willing to change my mind given the evidence.
If you do not contest my claim then the first question one has to ask is .... is the market random? then null hypothesis should be that it is; unless the evidence points to it not being random.

so how do we test for randomness? let us begin with asking this questions in the statistical sense.... for the data to be random it must follow a random distribution? do you agree? here are two distributions

1- fitted to the euro/usd named "DEXUSEU" from the Saint Lous FRED, this data is a total of 4227 euro logarithmic returns


here is the density of a randomly generated data of a sample size of 4227


if we do a QQ plot of euro.log , normally distributed data we can see the fit


My thoughts: the euro log density appears to have more peakedness and heavier tails, this however only shows that the volatility of the data is different not that it is or isn't predictable. so how do we do a test for randomness?

statistically we would see if there is any serial correlation (auto correlation) meaning: Autocorrelation, also known as serial correlation or cross-autocorrelation,[1] is the cross-correlation of a signal with itself at different points in time (that is what the cross stands for). Informally, it is the similarity between observations as a function of the time lag between them. It is a mathematical tool for finding repeating patterns, such as the presence of a periodic signal obscured by noise, or identifying the missing fundamental frequency in a signal implied by its harmonic frequencies. It is often used in signal processing for analyzing functions or series of values, such as time domain signals.-- wikipedia


here is the auto correlation of a random data


here is also the Partial Auto correlation function:
In time series analysis, the partial autocorrelation function (PACF) gives the partial correlation of a time series with its own lagged values, controlling for the values of the time series at all shorter lags. It contrasts with the autocorrelation function, which does not control for other lags.
This function plays an important role in data analyses aimed at identifying the extent of the lag in an autoregressive model. The use of this function was introduced as part of the Box–Jenkins approach to time series modelling, where by plotting the partial autocorrelative functions one could determine the appropriate lags p in an AR (p) model or in an extended ARIMA (p,d,q) model.

PACF of the Euro

PACF of random data


Conclusion: as you can see the Euro does not breach the confidence intervals, the data does not appear to be correlated which supports the null hypothesis of random data. here are the results of the correlation

Autocorrelations of series ‘x’, by lag
0 1 2 3 4 5 6 7 8 9 10 11
1.000 0.012 -0.013 0.000 0.035 -0.035 0.016 0.028 -0.002 -0.003 -0.019 0.006
12 13 14 15 16 17 18 19 20 21 22 23
0.013 0.018 -0.004 -0.012 -0.006 0.013 0.005 0.016 0.008 -0.012 0.011 -0.012
24 25 26 27 28 29 30 31 32 33 34 35
0.008 -0.004 0.007 0.002 0.005 0.017 -0.004 -0.002 -0.003 -0.008 -0.007 0.007
36
-0.011
> acf(rnorm(4227), plot=FALSE)
Autocorrelations of series ‘rnorm(4227)’, by lag
0 1 2 3 4 5 6 7 8 9 10 11
1.000 -0.029 0.034 -0.034 0.010 0.007 0.004 -0.006 -0.002 -0.024 0.009 0.000
12 13 14 15 16 17 18 19 20 21 22 23
-0.015 -0.020 0.006 -0.006 0.013 0.005 -0.003 0.001 0.010 0.002 0.014 -0.002
24 25 26 27 28 29 30 31 32 33 34 35
-0.006 0.011 -0.001 -0.008 -0.013 0.004 0.017 0.004 0.020 -0.012 0.014 0.002
36
-0.007

let's look for trend and volatility in the data the first one is euro, the second is a random normally distributed data




I can see a difference in the distribution, the upper red and lower red lines are at sigma one (standard deviations 1 and -1) the red line in the middle is the SMA of 22 = monthly, the yellow lines is the SMA 22*12= 264 which would represent a yearly smoothing, the blue line is to illustrate H=0

Brain sees a difference in the distribution although the monthly smoothing looks similar i can see either real or imagined evidence of trend in the data.

lets look at the Hurst Exponent

> hurstexp(x)
Simple R/S Hurst estimation: 0.5614027
Corrected R over S Hurst exponent: 0.5815521
Empirical Hurst exponent: 0.5223035
Corrected empirical Hurst exponent: 0.5014898
Theoretical Hurst exponent: 0.5211121

> hurstexp(rnorm(4227))
Simple R/S Hurst estimation: 0.5169296
Corrected R over S Hurst exponent: 0.5302739
Empirical Hurst exponent: 0.5114854
Corrected empirical Hurst exponent: 0.4897964
Theoretical Hurst exponent: 0.5211121

the Hurst exponent is another measure of randomness especially Persistence, positive where H>50 H<1 would indicate a trending component while H>0 H<50 would show mean reversion.
The Hurst exponent is used as a measure of long-term memory of time series. It relates to the autocorrelations of the time series, and the rate at which these decrease as the lag between pairs of values increases. Studies involving the Hurst exponent were originally developed in hydrology for the practical matter of determining optimum dam sizing for the Nile river's volatile rain and drought conditions that had been observed over a long period of time.[1][2] The name "Hurst exponent", or "Hurst coefficient", derives from Harold Edwin Hurst (1880–1978), who was the lead researcher in these studies; the use of the standard notation H for the coefficient relates to his name also.
In fractal geometry, the generalized Hurst exponent has been denoted by H or Hq in honor of both Harold Edwin Hurst and Ludwig Otto Hölder (1859–1937) by Benoît Mandelbrot (1924–2010).[3] H is directly related to fractal dimension, D, and is a measure of a data series' "mild" or "wild" randomness.[4]
The Hurst exponent is referred to as the "index of dependence" or "index of long-range dependence". It quantifies the relative tendency of a time series either to regress strongly to the mean or to cluster in a direction.[5] A value H in the range 0.5–1 indicates a time series with long-term positive autocorrelation, meaning both that a high value in the series will probably be followed by another high value and that the values a long time into the future will also tend to be high. A value in the range 0 – 0.5 indicates a time series with long-term switching between high and low values in adjacent pairs, meaning that a single high value will probably be followed by a low value and that the value after that will tend to be high, with this tendency to switch between high and low values lasting a long time into the future. A value of H=0.5 can indicate a completely uncorrelated series, but in fact it is the value applicable to series for which the autocorrelations at small time lags can be positive or negative but where the absolute values of the autocorrelations decay exponentially quickly to zero. This in contrast to the typically power law decay for the 0.5 < H < 1 and 0 < H < 0.5 cases.--wikipedia

finally we can evaluate the Runs Test for randomness


> runs.test(x,plot=TRUE)
Runs Test
data: x
statistic = 2.1203, runs = 2163, n1 = 2080, n2 = 2107, n = 4187, p-value =
0.03398
alternative hypothesis: nonrandomness

> runs.test(rnorm(4227))
Runs Test
data: rnorm(4227)
statistic = 1.3846, runs = 2159, n1 = 2113, n2 = 2113, n = 4226, p-value =
0.1662
alternative hypothesis: nonrandomness

as you can see the P-value on the data x =eur/usd appears be against the null hypothesis of randomness

to conclude my opinion, I claim that with the statistical tools at our disposal it is very difficult to tell the Eurusd data from a Random normally distributed data, therefore T/A alone would be similarly difficult to differentiate from a brownian motion monte carlo simulation and the real market. I claim that you cannot profit from T/A alone, hence the allegations of consistently profiting are highly likely to be false.

Can you tell the difference between Monte Carlo simulated data and the real market data?

i think Mandelbrot says basically the same things in his book on the Misnehavior of Markets? But anyways, what you say is convincing.

Wholeheartedly agree.

I haven't met a rich technician. Excluding, of course, technicians who sell their technical services and make a lot of money.

I personally don't believe in true randomness, and chance..... When a person say that something is random what he actually is saying is - "I don't understand this". That's the truth. We don't really understand how the world works!
I see that you have done your home work and you put some real effort to investigate in attempt to find the truth. But I think that you are digging in the wrong place.(like most traders) You will not find anything in the price history, because there is nothing there......
You know "Allegory of the Cave" ? Where the prisoners look at the shadows and think this is the real world......... To look only at historical prices in attempt to understand the market is exactly the same! You look at the shadows.... Prices are the end result, not the beginning or the cause of things. Each moment in the market is unique! Yes, the technical patterns repeat themselves but they do not predict the future. Only when some patterns appear in specific context, they give clues and maybe slight probability bias. But you have to think what is actually happening behind the scenes in each specific situation. You have to analyze the actions of the market participants....to understand what is the nature of the game..... I suggest you check the reflexivity theory, the negative/positive loop feedback examples, psychology and crowd behavior.
The only way to gain edge in the markets, is to live in the present as much as possible and analyze each situation and moment as separate event in time, fully independent of past or future.

I only use technical analysis. It works well. I am not rich, but trading gives an additional ice-cream money.

I think you complicate too much. Here is a proposition for a ridiculously simple technical system that should make money (as it makes sense). Timeframe and instrument acording to one's preferences. You use two indicators based on moving averages (!). Price and volume oscillator. Both operate using a difference of two moving averages of price or tick volume, respectively. The parameters of the oscillators should be set to the same value. One takes signals when the price oscillator gives you a signal and the volume oscillator is in the phase where it's increasing from a minimum (as tick volume activity usually leads price). SL below/above a recent swing, TP once price oscillator starts moving in the opposite direction. For intraday trading best to choose active sessions for the chosen instrument.

Cheers,
k

Gosh, this hurts my head.... and I have a science background! It's not that complicated!

On randomness: i do not subscribe to theories of randomness as applied to financial markets in general and forex markets in particular. I believe there is a reason for every bar on every forex chart , for every pair, for any time frame that was ever printed or will be printed in the future, Some are fundamental, some are technical,While it is true that the variables can be infinite and our ability to measure them in real time is limited because the mind of man is finite even with the aid of the fastest computers we cannot ascribe " randomness" to the condition. Example: Most everyone would agree that a croupier spinning a roulette wheel in a casino would produce random outcomes if not rigged.Really ?? If we knew the speed of the wheel, the force applied to both the spin and throw of the ball, the angle and arc of the throw and an infinitely large number of other variables we could determine its destination. Obviously we cannot know all that quickly enough to place a bet so we term it randomness. Therefore I conclude "randomness" is not a real fact but rather a state of mind caused by our inability to recognize and process large numbers of variables fast enough. Our problem as traders a and analysts is determine those salient variables that determine forex outcomes more accurately than those that do not and use them to trade.

Very important point!
In trading there is no need to understand how or why something works in order to make money. Just like you don't have to understand the complex technology inside your computer in order to use it......
To be profitable trader you just have to learn the price action patterns, learn how the market responds to news, fundamentals.... then based on that develop framework and strategy......and trade according to that strategy, and manage the risk. That's all.
If you do these things correctly, consistently and with discipline - the money will accumulate in your account almost effortlessly.
But for most people this process is impossible to achieve without real hard efforts. Psychology, cognitive biases and the social brainwashing prevents people from being successful traders. By default the human brain cannot deal effectively with probabilities, uncertainty, risk,. ....and exponential functions!
From very early age people learn to avoid risk and danger! Our whole society, and the education systems turns people into security/stability seekers.
And in the markets you have to do just the opposite! You have to take risk every day, deal with probabilities, embrace the uncertainty......all these are unnatural things and against the social norms. That's why it's so hard for people to beat the market.
And that's why 95% of traders look for shortcuts and look for ways to avoid the risk and the uncertainty. People hate to lose and naturally want to be right every time. And that's just not how this trading game works. Trying to avoiding the unavoidable leads to failure. In trading you have to lose in order to make money. It's a paradox! Very hard thing to overcome psychologically for most people.

Well put folks. I would add it's basic human nature to make that which isn't understood complicated/complex. It's easier to accept our lack of understanding when we tell ourselves it's so hard. It gives us a pass to say "no way we can understand it" and thus releases us from taking the time and doing the work needed to figure it out. It's a comfort thing.

IMHO, those who dedicate themselves to that work will inevitably find the work to figure out how easy something really is, was actually the hardest part.

very interesting. may i ask you a chart example please? tks

The above is very true.

Though the problem I have with technical indicators like RSI, MAC D etc. is a signal on one time frame normally has an opposite signal on another timeframe so its nearly always 50/50.

The edge is in the money management with these strategies and that's not good enough for me.



On another note I always wonder how quantum computing will effect the markets?

I can't really speak to such indicators as I simply don't use them. However, you must realize the market as related to time frames is fractal in nature. Smaller TF's are simply a smaller version of the larger ones. So if one were to stick to just one time frame, I can't help but wonder if such indicators wouldn't be more accurate than 50/50. I think they would be. So going by whatever they say on one TF, would be no different than what they say on another. Only the "time" it takes to move them a given distance changes. Obviously "comparing" them across multiple TF's will result in an "averaged" or, closer to 50/50 result.

Price is price regardless of the TF being viewed. The "structure" of price movement is quite similar. The "distance" of those structures is what changes across TF's.

And of course, in any business, money management is key to overall profit or loss. Cash flow is just that. It constantly flows in and out. The desired result is of course obvious and proper management will get you there.

surely mathematics and computer logic would never = human logic
its all man made by many minds , theres no way that would work I'm confused as to why you think the market can be measured in this fashion ,
apply mathematics to the human mind your get nothing ?as a result
which is what you did

Thanks for starting this thread as your content is of interest to me.. I hope you don't mind that one of the "bullshit" spreaders has arrived, because I have totally different beliefs than you and I support those views by results.

For my introduction, I would like to say I have zero background in statistical analysis unlike yourself. I do feel this gives me a great basis for my beliefs because my views are not formed based on scientific formula and calculations, but by observations and practice. I worked as a coal miner for 37 years. During this time, I invested and traded in equities markets where I began studying patterns. As a manager for my company, I was compensated well. I retired at the age of 55, walked away from a job that would have paid me a guaranteed income of 1 million dollars or more, had I worked till I was 62. At that time, almost 4 years ago, I became a full time forex traders. Forex does not provide me with "ice cream" money, it provides me with serious money from my trading, serious enough that I knew I could walk away from that guaranteed income. But I should say I didn't leave my job and decide to do this for the money, I left my job to be a caregiver to my wife, a cancer survivor of 11 years now, who is in falling health. Your convictions in your beliefs allow you to trade for "ice cream" money, while mine allows me to replace lost income from a job, and to stay home and take care of my wife. Enough of that....

I have read your theories here as well as in other threads and disagree that the market is random.

As with all computations, you must have a clean data set, free of corruption, in order to have clean results. It is my view that every computation that you have run is based on corrupt and false data, leading to flawed output. Or do you disagree about needing clean data?

I replied to a thread here about price prediction back in 2012, where I presented my views as well as a chart of the EURUSD pair with my "price predictions" in the open for all to see. While I do not attempt to predict specific price points, say to a pip, I do predict price direction very well, to the point that it will follow the maps that I draw.

I have heard so many people try and explain to me that it isn't possible and use rational means in doing so, but as of yet, none can explain away the fact that I do it, or the fact that I have been able to guide others in doing the same thing. If I and others can do this, then how can the market be random in any fashion?

I have posted a map for EURJPY in a thread here as well, the 2nd public price map that I have thrown out for anybody to follow. I have also recently created a private price map, for AUDJPY that I am trading for serious money. This map was also sent to 2 members here at FF and you know who one of them was. Time again, will prove me right. It is up to you and your computations to prove it wrong, not rationalize why it shouldn't or can't work when it does.

I leave this post with 3 questions....

1. If computations are based on corrupt data, can the results of those computations be trusted to be true?
2. What are the probabilities of a person or persons projecting price as I do, not once accurately, but in time, three times accurately.?
3. It is generally accepted that 95% or more, of forex traders lose money. If the market were indeed random, then why would the number not be closer to a 50/50 ratio, over time? This would be closer to the normal probability results of a random coin toss with 2 outcomes would it not?


-CJ-

Agreed. We do in the blink of an eye what the most powerful computers on the planet can not. Facial recognition is in it's infancy in the world of science but we do it in an instant. Pattern recognition is how we do it and we do it well, not only on faces but with all patterns.

-CJ-

exactly Captain ,I've thought about this recently and your probably think I'm crazy ,but how does a chart have elements of life in it ,like fib ratios ,fractal nature throughout tfs
I've come to the conclusion that all the connected humans creating a chart
result in this , naturally

Assume that there are 100 super smart brilliant people in a room with very advanced degrees.

All of them are asked the question - are markets random?

Half of them say yes and other half say no with concrete proof.

Where do stand then?

That's how this thread and many many other threads asking this very question are going to be...for now and in future...

An alternative question might be "Can you tell the predicted price movement from the real price movement?"

The original chart was posted in 2012, here at FF. The results follow the predicted path of price movement proving that price can indeed be predicted, and not only predicted , predicted with great accuracy.

-CJ-

Alright, time for a serious conversation on the subject.

If we accept that price is random and the trends in price data are no different from the trends that may also occur in random data then people who profit from trading are simply lucky survivors of a random process.

Some people are lucky to be born smart enough to realize that markets cannot be solved. Others are even smarter and were born lucky.