Thanks, I appreciate the feedback. The one limitation I have right now is I cannot check stats of intra-day movements such as it will reach price X only after it hits price Y during the day. However I am able to pull other things to make the prediction more complete, such as:
80% chance GOOG will have day High of at least Open + 10.00 if maximum day Low is Open - 8.00.
So the trade would put a stop at Open-8.01 (for that 20% chance).
Besides these type of predictions are there other ones that you think are valuable?
Currently the neural network I am building for predictions is also able to have predictions from multiple stocks at the same time and also further in the future.
Example:
If rimm.Close drops for 3 consecutive days, then 80% chance aapl.High within the next 3 days will be at least 5% higher than the aapl.Close from yesterday.
So, say yesterday is Day 0 (the latest trading day):
Day -2: rimm.Close = 9.99
Day -1: rimm.Close = 9.90
Day 0: rimm.Close = 9.70 ; appl.Close=550.00
Then Prediction is:
Either on Day 1, Day 2, or Day 3 (within the future 3 days), appl.High will be at least 577.50.
But to your point, I would also need to put in a stop conditions (conditions that would violate the prediction).
Thanks.
-Rob
80% chance GOOG will have day High of at least Open + 10.00 if maximum day Low is Open - 8.00.
So the trade would put a stop at Open-8.01 (for that 20% chance).
Besides these type of predictions are there other ones that you think are valuable?
Currently the neural network I am building for predictions is also able to have predictions from multiple stocks at the same time and also further in the future.
Example:
If rimm.Close drops for 3 consecutive days, then 80% chance aapl.High within the next 3 days will be at least 5% higher than the aapl.Close from yesterday.
So, say yesterday is Day 0 (the latest trading day):
Day -2: rimm.Close = 9.99
Day -1: rimm.Close = 9.90
Day 0: rimm.Close = 9.70 ; appl.Close=550.00
Then Prediction is:
Either on Day 1, Day 2, or Day 3 (within the future 3 days), appl.High will be at least 577.50.
But to your point, I would also need to put in a stop conditions (conditions that would violate the prediction).
Thanks.
-Rob
