Linear discriminant analysis (LDA) is a very common dimensionality reduction technique for classification problems. Like kohonen maps in prior article if you have high-dimensional data (i.e. with a large number of attributes or variables) from which you wish to classify observations, LDA will help you transform your data so as to make the classes as distinct as possible. LDA is very much like the techniques PCA, QDA, & ANOVA; and ...
Probably, every trader's dream is an indicator able to adapt to the current market situation, define flat and trend segments and consider relevant price changes. Conventional technical indicators use constant ratios when handling input signals. These ratios do not depend in any way on the characteristics of the input signal and its changes over time. Adaptive indicators are distinguished by the presence of feedback between the values of ...
In a previous article Development of a trading EA from scratch (Part 29), we have removed Chart Trade from the EA. Previously, we did the same to some other thinks, like Volume At Price and Times & Trade, to improve the EA performance and reliability. By removing Chart Trade from the EA, we are left with only the basic order system. Although this may seem insufficient to some users, the EA can actually do all the work. But there are people who like to enter and exit trades in the market, but they ...
We continue to study model training algorithms. All the previously considered algorithms used an analytical method for determining the direction and strength of changes in model parameters during the learning process. While the methods to solve these problems turn to be inefficient. In such cases, we resort to evolutionary optimization methods. more... --------------------- Neural networks made easy ...
We continue to explore reinforcement learning methods. In previous articles we discussed methods for approximating the Q-learning Reward function and the policy gradient function learning. Each method has its own advantages and disadvantages. It would be great to use the maximum of their advantages when building and training models. When trying to find methods minimizing the shortcomings of the algorithms used, we often try to build certain conglomerates ...
Recent Comments