investment behaviours can be automated and all the data is available online, so all strategies can be played out.
types of implementations: supervised neural network, quantitative (based on decision making criteria and patterns), automatic (like a recipe), genetic (unsupervised ML).
they need to be tested in order to evaluate the accuracy of their predictions. Consequently, algorithms that focus on intra-day trading are easier to be tested and adapted. Previous data can also be used to back-test the bots.
positions can’t be too small since commissions may eat big chunks of the profits.
the matter requires more investigation. Buzzwords to read about: LSTM, single layer perceptors, Pine script.
Oil painting is tougher than expected and no, gin cannot be used as a paint thinner.
Can a console game be opened and resealed in a professional looking way?
