Introduction & Concept

Even if you had the perfect signal to open a position, how will you maintain that position or multiple positions?

Main questions with a perfect prediction

• At which point would you set a TAKE PROFIT order? Or a STOP LOSS?
• What ratio would you use for open and close and why?
• Will you wait for a close signal as well? What if you miss that close signal window?
• What is the best trading strategy for 1 stock or 10 stocks?
• What is the best budget for a stock?
• What is the best scenario of all of the above for $100K or $1M?
• What is the best set of parameters for trading if your preferences limit volume or price range?
• What should you do if you only trade 50 stocks and want to see the best strategy to manage that portfolio with and prediction in between?

Simulation vs experiment

A simulator runs using a set of parameters that characterize trading strategies. Each run is called an iteration. The result of an iteration is an experiment.

Simulation domains

Tickbooster simulator always runs 3 simulations at each iteration: prediction, no prediction and SP500. It means we want to see how does our prediction performs when used during trading, when not used during trading and when we only buy SP500 at the beginning of the period and sell everything on close.

Trading strategy

So with the perfect prediction, you’ll still need to see how to trade symbols if you used that prediction signal for opening a position. Tickbooster is for traders who do not want to sit all day in front of their desks waiting for signals to come and go. Tickbooster is for traders who fire and forget their trades because they are sure that the prediction signal they used for picking and the trading strategy helps them set purchase limit, stop loss and take profit parameters well that is in line with their personal budget and risk taking preferences.

Strategy simulator overview

Tickbooster simulates trading based on a set of parameters like take profit, stop loss, expiration sell, sell on any profit and many more. Then it runs the simulation over many iterations so it’s sure that you’ll not only receive one particular result but an aggregate of many experiments. So when you go to the strategy designer, set your preferences and hit run simulation the following happens:

Simulation algorithm

1. Pick a past date range of 2-3 months
2. For each iteration generate a random seed and:
   1. Pick a random set of symbols for a given total and per symbol budget
   2. Do a day-by-day trade simulation from the beginning to the end, on each day:
      1. Open a position if there is a symbol with proper prediction strength and there is still budget available.
      2. Close a position if there is a symbol that meets closing criterias like expiration, take profit, stop loss etc.
   3. At the end of the simulation period close remaining positions
3. Aggregate each independent iteration results