Application of mathematical analysis and higher mathematics - page 2

New comment
 
Reshetov писал (а):
To begin with, look in a calculus textbook and ask what the difference between interpolation and extrapolation is.

I'll be sure to refresh my memory :-) Although I think I made the right point ....
 
Cronex писал (а):
Reshetov wrote (a):
For a start you need to look in a calculus textbook and ask what the difference between interpolation and extrapolation is.

I'll be sure to refresh my memory :-) Although I think I got it right ....
My robot is very interesting to me, but I've got some dummies' questions and I'd really like to talk to you one-on-one on this subject (if you don't mind, of course)!
 
krit писал (а):
 My robot SAR trading v2 is new to me and I'm very interested in it. I'm new in it (rather dummies).Very interesting to me your trading robot, but I have some dummies questions and would love to talk to you one-on-one on the subject! (If you do not mind!!!) ICQ 293-387-521 (please).
Krit, thank you for your interest :-)
I dont use aske on principled reasons, but i can contact you by phone or by e-mail.
Leave your coordinates.
 
Cronex писал (а):
Here's an idea for example: calculate a probable vector for a graph, restore it to a target by quadratic interpolation and then calculate the divergence from the real behaviour to assess the direction and quality of the trend.

If you can, answer the questions:
1. How do I calculate the likely development vector ?
2. How do I calculate the target?

3. Do you think that if the probability of prediction (vector of development) > 50%,
then the TS built on this technology is doomed to success?
Say, if this vector will be built all the time with the probability of 70%,
then the ratio of profitable and losing operations must be = 7/3?
 
SK. писал (а):

If possible, answer the questions:
1. How to calculate the likely vector of development ?
2. How do you define the target?

3. Do you think that if the probability of prediction (vector of development) > 50%,
then the TS built on this technology is doomed to success?
Say, if this vector will be built all the time with the probability of 70%,
then the ratio of profitable and losing operations must be = 7/3?
Please accept one position, it is not a statement, but a supposition and a purely personal opinion.
1. Well you will probably laugh here, but I tried to apply thermodynamics, in particular the behavior of industrial steam generators. Perhaps I am reinventing the wheel, but the SAR indicator determines this dynamics very closely if you work with it on daily charts when accounting for trading volumes. In addition, for additional confirmation of the vector direction, you can use the quadratic interpolation of the previous period, projecting half a period ahead and extrapolating to the target.
2. I think to define the target we may apply the basic part of the wave theory (in particular the theory of convexity and continuity in the point) plus Fibo levels and accept the probability of development not in our direction; in my own development the levels 0,382,0,618 and acceptance of probability of failure are very useful.
3.No I do not think so, because of many other factors that influence the result of the deal (stop technology, cutting the double losses, resource management, and the state of the economy, etc.). Any profitable position can be spoiled :-)
For a successfully predicted vector, let's take 70/30 for D1, the ratio of profitable trades can only theoretically be 70/30 since prices correlate both ways during the day, when the general direction is correct. The test accounts I tested on are much more interesting, from 62/38 to 96/4 during the week... that's when you start thinking about Fibo.
 
And I don't think that all that high maths gives you much results. It is not because I am not good at it, on the contrary. In the past, when the whole stock market was almost "manual", i.e. all operations were done manually with a broker's fingers on a live exchange, there were no problems. There were waves, which some people even described with vibrational functions. And what do we have now? The market reacts instantly. To what, I ask you? To financial events that should not be predicted by higher mathematics but by financial analysis. All you need is a bright mind. If our experts learned to read news, and assess their importance correctly, that would be the key to success. Hence the moral: you need a team of people who collect the news, and put them on the server in a readable form for experts, and they sell themselves. Connection to the server has to be paid for. Although, firms that give recommendations, also scattered a decent amount. There are even serious...
 
GeMeL:
And I don't think that all that high maths gives you much results. It is not because I am not good at it, on the contrary. In the past, when the whole stock market was almost "manual", i.e. all operations were done manually with a broker's fingers on a live exchange, there were no problems. There were waves, which some even described with oscillatory functions. And what do we have now? The market reacts instantly. To what, I ask you? To financial events that should not be predicted by higher mathematics but by financial analysis. All you need is a bright mind. If our experts learned to read news, and assess their importance correctly, that would be the key to success. Hence the moral: you need a team of people who collect the news, and put them on the server in a readable form for experts, and they sell themselves. Connection to the server has to be paid for. Although, firms that give recommendations, also scattered a decent amount. There are some serious...
Exactly!
 
GeMeL:
And I don't think that all that high maths gives you much results. It is not because I am not good at it, on the contrary. In the past, when the whole stock market was almost "manual", i.e. all operations were done manually with a broker's fingers on a live exchange, there were no problems. There were waves, which some even described with oscillatory functions. And what do we have now? The market reacts instantly. To what, I ask you? To financial events that should not be predicted by higher mathematics but by financial analysis. All you need is a bright mind. If our experts learned to read news, and assess their importance correctly, that would be the key to success. Hence the moral: you need a team of people who collect the news, and put them on the server in a readable form for experts, and they sell themselves. Connection to the server has to be paid for. Although, firms that give recommendations, also scattered a decent amount. There are even some serious ...
I do not agree completely. I don't agree. First, how do you expect to teach an expert to assess news, and secondly: when you could trade on crossovers of averages, and now everybody has computers and it just does not work, that is, the market has become more knowledge-intensive, that's all.
 
To start with, it's a good idea to answer questions like:

1. What is the market?
2. What does it consist of?
3. What is traded on the market?
4. What is a price?
5. Why does the price move?
6. What influences the price movement?

For simplicity, it is better to take the stock market (shares), not Forex.

In this case, the question how to trade and what requires mathematical analysis may be answered,
If you want a mathematical analysis or a fundamental analysis, then (a)).
 
How do you distinguish a trending movement from a sideways movement and switch to the appropriate type of trade, i.e. a post-trend or counter-trend? This can be determined with the Perceptron oscillator - a single layer neural network. Its description and source can be found HERE

The interpretation of the oscillator readings is primitive:

A value above zero - open long or take profit on a short one
A value below zero - to open a short one or fix profit on a long one

Disadvantages of the oscillator:

1. It must be taught on real (not on demo accounts) quotes of a security
2. It makes a mistake when an up-trend or down-trend sharply changes direction or turns into a sideways trend. Moving to a sideways trend from a vertical 50/50
3. It is incorrect if a sideward trend changes to a vertical 50/50

If trends are prolonged, regardless of whether they are sideways or vertical, then directional changes (volatility) are rare, or rather only at the beginning of the trend and errors are unlikely.
123456789...13
New comment