Strategic foresight systems - page 15
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
Let's be clear, so that we understand each other. In search engines you can find many different interpretations of the term trend. What would we consider a trend?
Yes you are right, we need to clarify the definitions. All further IMHO:
You should agree that one candle on a daily TF is not a trend, but if this BP is considered on an hourly TF then a trend with all trend characteristics of impulses and corrections will undoubtedly be observed here. Or will an impulse or a correction separately be considered as a trend?
I also introduce a "trading" definition of a trend as finding the best (optimal) TP take based on the identified pattern. Additionally, this definition(trading time):
For reference and better understanding with my colleagues, I have introduced identification of my current trend and I will further support this initiative. The identification includes:
It will be easy to compare my trend with your "current" trend. I think in most cases they will roughly coincide.
PS: To put it philosophically, a trend is when you just know how the price will develop. Still, a trend of one or three days length is not very common, rather rare and is a peculiarity of its classification. It is not worth paying attention to.
We haven't got a clear definition of the trend. Come on, we'll get used to it.
And this is how clearly we closed the month's observable toolkit. They zeroed out almost perfectly. Maybe I have an error in my calculations, I need to check it.
We haven't got a clear definition of the trend. Come on, let's get used to it.
My understanding is that the definition is more than specific (i.e. several very specific time series models are used). For more understanding it would be interesting to hear your definition.
And here's how clearly the month's observable toolkit closed. Zeroed out almost perfectly. I may have an error in my calculations, need to check.
I hope I can get the astrolabe together in a couple of days...
I like this interpretation of the trend the best.
Each successive dip should be higher than the previous one and each successive peak should also be higher than the previous one for an uptrend. Each subsequent node should be lower than the previous one, and each subsequent dip should also be lower than the previous one for a downtrend.
Of course it should only be applied on the timeframe in question. Each stretch of up or down price in the trend can and should be treated as a lower order trace.
I don't use 33, I have a different approach to it.
I hope to get the astrolabe together in a couple of days...
Everyone is already waiting for it :), but better to do it more reliably than faster.
I like this interpretation of the trend the best.
Each successive dip should be higher than the previous one and each successive peak should also be higher than the previous one for an uptrend. Each subsequent node should be lower than the previous one, and each subsequent dip should also be lower than the previous one for a downtrend.
Of course it should only be applied on the timeframe in question. Each stretch of upward or downward price in the trend can and should be considered as a lower order trend.
...
I see, you have the approach of a real artist and I have the approach of a former one :o) To your definition you need to somehow attach a predictive assessment of future trends.
Everybody's been waiting for it :),
:o) I don't think so, who needs a system in debugging except its maker :o)
But it's better to make it more reliable than faster.
But there are problems with reliability, and very tricky, quite unexpected for me. To build the market as a single model (so to say - theory of everything, i.e. M-theory:o)))) ) I used the principles of multivariate analysis. And it resulted in prediction of a quote taking into account influencing factors (other correlated quotes and some other stuff) increased the forecast accuracy 2.4 times more for each quote, than in case I would have predicted it separately.
But out of 14 (and a couple of other external influencing factors I tried) about 10 of them seem to be making good predictions, but some of them start "stochastically" jumping. I.e. "yesterday's" is very different from "today's". That's what confuses me.
So I'm taking another time out, probably until Monday. If I don't figure out what's wrong, I'll roll back to the old model and meditate on this one.
I see, you have the approach of a real artist and I have the approach of an ex-artist :o) To your definition you need to somehow attach a predictive assessment of future trends.
A daily trend is nothing more than an uptrend or downtrend if you look at the chart from the perspective of a weekly timeframe, so a line chart has no problem with that.
In this chart it is the other way around. We look from the 4hour timeframe to the projection of the daily timeframe. No difficulty for real artists.
By creating a combination of different TFs it is possible to see with a high degree of accuracy the upcoming behavior of almost any instrument. I don't have skills like yours yet, so I am watching the project with interest.
An artist's view from a higher viewpoint. This is the picture that emerges
picture.
A daily trend is nothing more than an uptrend or downtrend in terms of a weekly timeframe (TF), so a line chart has no problem doing that.
In this chart it is the other way around. We look from the 4hour timeframe to the projection of the daily timeframe. No difficulty for real artists.
By creating a combination of different TFs it is possible to see with a high degree of accuracy the upcoming behavior of almost any instrument. I don't have skills like yours yet, so I am watching the project with interest.
An artist's view from a higher viewpoint. You get such a picture
picture.
The script pumps up the history, but only if the appropriate history files have already been created, i.e. charts with the appropriate TF have been opened at least once and the terminal has been reloaded, I think that you, Sergei, will definitely have the script working correctly
All that's left now is to make the forecasts! :D
I'm sorry to bother you, I have a question. I have a suspicious behavior. Yesterday I downloaded all of the history at once and after the start the indicator for loading the history started "blinking" and loading at a speed of about 1 bar every few minutes. Watched the process for a few hours, thought it would "grunt", didn't wait for it to finish. My internet speed is very good, i doubt it's the reason.
I'll try again today.
Igor, sorry to interrupt - I have a question. The script has started to behave strangely. Yesterday, when I started it, all the history was loaded, but after start the history loading indicator started "blinking" and loading at a rate of about 1 bar in a few minutes. Watched the process for a few hours, thought it would "grunt", didn't wait for it to finish. The internet speed is very good, i doubt it's the reason.
I'll try again today.
no predictions for today (next week)?