Just to give an idea of what I'm doing:

I'm doing a project with financial data (tick data) and I'm trying to create a way make a model learn what happens before a breakout.

Usually there is a disproportionate ratio between buyers and sellers and the interval between the orders usually is shorter (people are buying or selling faster so they don't miss out). This happens a couple of times every day in the asset that I trade in Brazil.

So my question is: How do I train "mini datasets" and then put all of them together so the model learns what a good breakout looks like. I want work initially with these two variables:

  1. time interval between orders and
  2. order volume

Any tips on which algorithms or subjects that I can study to do this?

Here is the dataset that I have. It also include the broker which the order came from. I have 2 years of data like that and I know when it breaks out. Any tips?

enter image description here

  • 2
    $\begingroup$ Thanks for your contribution. Please try to formulate a more concrete question. Asking for tips is discouraged in this forum. Questions that allow for a clear answer are preferred over question that may lead to open-ended discussions. $\endgroup$
    – oW_
    Jun 29 '19 at 7:04