First sacred cow: "If the trend started, it will continue" - page 54
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Ребята видят феноменологию. Почему - не важно.
После А, следует В. Является ли А причиной В ? Совершенно не факт. Но имеет место быть. Я об этом.
If you, a 21st century citizen, would rather believe in tambourine dancing, that is your right. But the big money is made by those who know "why".
In order to be successful, you have to know whether A is the cause of B. Otherwise you risk running into unaccounted C which will ruin everything. All the stats are about this very issue - the real relationship between A and B.
Just posted in the correlation thread - more imports of lemons "leads" to fewer road deaths - such is the "phenomenology". Finding a trend in the financial markets is the same as trying to reduce deaths by increasing imports of fresh lemons.
так и будет: средняя температура будет повышаться от зимы к лету и наоборот и необязательно линейно как у вас в формуле.
This is seasonality from winter to summer, not a trend. The average winter temperature is always the same, i.e. there is no trend. If you step back a little further from the graph, you can see that the average temperature has been rising for the last XXX thousand years. Is there a trend? If you go even farther away, you can see that the average temperature has been fluctuating in a strictly horizontal channel for the last XXX million years, i.e. there is no trend here either, it is seasonal. Even in such a seemingly simple matter as average temperatures not everything is clear-cut. But there are answers to "why" seasonality.
But with financial markets there are no answers.
This is seasonality from winter to summer, not a trend. The average winter temperature is always the same, i.e. there is no trend. If you step back a little further from the graph, you can see that the average temperature has been rising for the last XXX thousand years. Is there a trend? If you go even farther away, you can see that the average temperature has been fluctuating in a strictly horizontal channel for the last XXX million years, i.e. there is no trend here either, but seasonality. Even in such a seemingly simple matter as average temperatures not everything is clear-cut. But there are answers to "why" seasonality.
And with financial markets there are no answers.
Trends in the sense you put the word, are probably a very rare phenomenon in financial markets. Perhaps it is only the long-term growth of the US indices, even when inflation is taken into account. But even that may come to an end once and for all(((.
Perhaps another definition of the trend is softer: probabilistic and short-lived. The trend is likely to continue for some time. It will not necessarily end after this time, there is simply not enough information to go far enough. If there is statistically proven profitability within this definition of trend and trendiness, then why not? All of these definitions and theories are non-universal and are not true or false, but differ only in their usefulness in a particular application.
Therefore, the main thing within this thread is to answer the question: which definition of a trend and trendiness has a practical use in the specific market. And it is even possible why this definition is useful.
Z.I. Although the topic-starter's formulation "If a trend started, then it will continue", implies that trends are short-lived and may replace one another, rather than an eternal drift in one direction. The way to identify the start and end of such trends comes to the fore.
З.Ы. Хотя и в формулировки топик-стартера "Если тренд начался, то он продолжится", подразумевается что тренды недолговечны и могут сменять друг друга, а не вечный снос в одну сторону. На первый план выходит способ идентификации начала и конца таких трендов.
You can call black white and white black. But if the trend has not continued, it is not a trend.
You (most people) continually ignore the main point in finding a trend - the "why". Why would there be a trend on the chart? Without the answer to that question, any construction is just random. "The way to identify the beginning and end of such" randomness comes to the fore. I have an answer - by chance, strictly by randomizer.
You can call black white and white black. But if the trend has not continued, it is not a trend.
You (most people) continually ignore the main point in finding a trend - the "why". Why would there be a trend on the chart? Without the answer to that question, any construction is just random. "The way to identify the beginning and end of such" randomness comes to the fore. I have an answer - by chance, strictly by randomizer.
I've always written that the answer to that question is paramount. You can on a spider, check it out.)) There are a number of options conventionally divided into investment and speculative (and reasons - exogenous and endogenous), but they all boil down exactly to local drives, not perpetual trends. These local drives continue temporarily, but the main thing here is timing - the correspondence to the processes that produce this drive. I have a negative attitude to trend following on FX (it was also discussed on the spider), but the use of trending is much wider than trend following. For example here is the latest discussion of the subject https://www.mql5.com/go?link=http://forex.kbpauk.ru/showflat.php/Cat/0/Number/260749/page/0/fpart/1/vc/1. You can find more detailed information in e.g. Neo's post about the sentiment in the same place. Do you want to discuss it here?
есть определенные моменты времени (если принмать что время линейно) наступает критическая точка и человек принимает решение о покупке или продаже тем саммым подталкивая цену двигаться замедляться и ускаряться тому процессу который в данный момент соотвествует тенденции - это и будет тренд..
If you take such a point on the hourly chart, for example, and then look at it on the minutes, you will see that while the trader or TS was picking his nose long on the hourly chart, the market reversal started long ago on the minutes.
Already for the 101st time, draw a random walk graph x(i) = x(i+1) + e(i), where e(i) is any stationary noise, and look at it for a long time and think. There will be "rise - fall", but no trends.
You should never forget what BP is generated by. If by a random number generator, that's one thing. Our series is spawned by a crowd that, for some reason, goes in step and in the same direction at the same time. This is what the initial process is, which we then measure, then input the squared, then make stationary, but never ask ourselves the question - what is the confidence interval at each point of our measurement? And going to stationarity, forgetting the confidence interval, we come up with a forward test.
You can call black white and white black. But if the trend has not continued, it is not a trend.
You (most people) continually ignore the main point in finding a trend - the "why". Why would there be a trend on the chart? Without the answer to that question, any construction is just random. "The way to identify the beginning and end of such" randomness comes to the fore. I have an answer - by chance, strictly by randomizer.
At the heart of our BPs is economics. If you build a new factory with a new product that is in demand, the market will see a rising trend in its share price. This is an objective reality that various analysts out there are trying to screw up (for personal gain, it seems to me). If it is a powerful plant and its production is for export, it can also give a trend in currency. Example, the impact of gas and oil supplies on the euro-ruble and dollar-ruble pair over the last year. Forget the formulas and never forget the real, practical source of trends in the form of watches, shorts and jeans...
Ребята видят феноменологию. Почему - не важно.
После А, следует В. Является ли А причиной В ? Совершенно не факт. Но имеет место быть. Я об этом.
I agree with this. If short-term trends can be detected on a random ramble, then they can be used in "trading" on that ramble. All the talk about whether they can be there or not (unless there should be a condition in the trend definition about their causes) makes no practical sense. The market may have some reason for such trends to occur, but because we do not know about them, we can perceive it as a random process (in many ways). But it does not prevent from profitable trading. So, there is no obstacle to it on random wandering either.
Если тебе, жителю 21 века, больше хочется верить в пляски с бубном, то это твоё право. Но большие деньги всё-таки зарабатывают те, кто знает "почему".
Для успешного использования обязательно надо знать является ли А причиной В. В противном случае рискуешь нарваться на неучтённый С, который всё испортит. Вся статистика посвящена именно этому вопросу - реальные отношения между А и В.
Только что постил в топик про корреляцию - увеличение импорта лимонов "приводит" к снижению смертей на дорогах - такая вот "феноменология". Поиск тренда на фин.рынках - это тоже самое как попытка снизить смертность путём увеличения импорта свежих лимонов.
Situations like the one in the example arise precisely from an excessive desire to know "so why?
1. Why do you care?
2. Have you yet to realise that market information is being tucked away for "suckers" to get them confused and flush?