Entry Signal - Profit per Bar
Author: hasenhas
Creation Date: 5/14/2019 10:06 AM
profile picture

hasenhas

#1
It could be important, determining the profit per bar (EOD) of an entry signal.

So you can find out the power of an specific entry signal. Normaly - if you have an good entry - the profit decreases day by day. For example on day 1 a good signal makes the most profit. Reducing it's power on day 2, 3, 4 and so on. Till there is no profit any more.

Is there a key figure that shows the entry signal power on daily base?

Thanks!
profile picture

Eugene

#2
Good point. The Trade Life visualizer from MS123 Performance Visualizers answers just this question (and more): is there a tendency of winners to stall after some point in time etc?
profile picture

superticker

#3
If you're looking for the entry signal power after the fact, the Trade Life visualizer mentioned in Post #2 is an option. But if you're trying to solve for this signal power in real-time (like within your strategy code) to determine when to buy, then there are some complications. So the process we are describing here is a first-order system, and we need to solve for the coefficient of the first derivative term. But to do this reliably requires about 4 or 5 bars, but after that much time has passed the opportunity to buy has probably gone. That's the catch-22 with any real-time solution. :-(

What I might suggest is to use the Kalman indicator (Yes, Wealth-Lab has one.) to solve your first-order system in real-time the "best it can," then use the slope (i.e. price "gain" in dollars/bar) of the most recent bar of the Kalman as your buy signal if it meets some threshold goal.

Also, not every rise in stock price is a first-ordered process (although many are), so you'll also want to test if the price behavior follows the Kalman indicator as expected. The simplest way to do that is by computing an R² correlation (goodness-of-fit) between the observed price and the Kalman price. If R² is better than 0.85, you're probably good to go. The problem is your R² calculation may not reach 0.85 after 3 bars because there's just not that much certainly in that short of time. It may reach 0.85 by 5 bars, but then you missed your opportunity to buy. So the catch-22 will always be with you regardless of your approach; you'll simply need to take a calculated risk based of the R²ed certainty for that buying moment.
This website uses cookies to improve your experience. We'll assume you're ok with that, but you can opt-out if you wish (Read more).