Deep Bollinger Bands
- Indicateurs
- Marat Sultanov
- Version: 6.19
- Mise à jour: 5 juillet 2023
- Activations: 20
Every indicator has its advantages and disadvantages. Trending ones show good signals during a trend, but lag during a flat. Flat ones thrive in the flat, but die off as soon as a trend comes. All this would not be a problem, if it was easy to predict when a flat changes to a trend and when a trend changes to a flat, but in practice it is an extremely serious task.
What if you develop such an algorithm, which could eliminate an indicator's flaws and enhance its strengths? What if such an algorithm could improve the operation of a trend indicator during a flat, but at the same time increase the efficiency during a trend, and fix the signals of a flat indicator during a trend and perfect them during a flat?
Deep is the algorithm that enhances the advantages of an indicator and reduces its disadvantages. This algorithm improves indicator quality as a result of combining neural networks with indicator algorithm.
About the Deep algorithm
The specified task, due to its non-reproducibility and complex regularity, are best soled by non-linear algorithms. Neural networks are a very versatile and powerful tool, which allows to achieve a lot, if the task is set correctly and the solution is formed reasonable. The algorithm is based on this tool.
This algorithm calculates the optimal periods and deviation of the Bollinger Bands, depending on the state of the trend and the flat. The algorithm trains the neural network to predict the optimal Bollinger Bands for each market condition, so the indicator adapts to the flat and trend, eliminating the shortcomings of the basic algorithm. In the flat condition, the neural network adapts the indicator in the direction of the flat, in the trending condition in the direction of the trend.
The neural network architecture and the heuristic search algorithm were designed specifically for the optimal solution of the task. For convenience, there are Presets of the neural network structure, as well as all the necessary parameters for fine-tuning the neural network.
Advantages
- To speed up calculations, you can use a graphics card (GPU), which greatly improves performance and results. In addition, the indicator can be calculated asynchronously using all the cores of your CPU and GPU.
- The algorithm has a self-learning system. This is an adaptive system that helps to learn flexibly all the time using a neural network so that the algorithm always keeps up with all market changes.
- Graphical control panel. Using the panel directly on the chart, you can flexibly control and manage the learning system.
- Parameters for fine-tuning all functions.
