Granted this is true. How are you going to take advantage of these opportunities?
what can be predicted and what cannot be in the market?
- Thread starter mute9003
- Start date
Granted this is true. How are you going to take advantage of these opportunities?
Over the last 5 years the largest day down in July (net change) was - 1,2 % on ES / SPX. Currently, we're down - 1,6 %.
My predictive model suggests a very high probability of trading back inside and closing inside yesterday's range in today's RTH session, but can't rule out a dip on the Open. Say within the 1st 30 minutes.
Unless today is one of those days...
Yesterday's Low was 4320,25. Today's High was 4322,75. Check.
Didn't get the Close inside the range (4.75 points shy), but not too shabby, IMO. Enough to make a buck.
If in the stock market it was possible to predict even a small part, all psychics and predictors would be fabulously rich. Unfortunately or fortunately, one can only guess price movements based on observations. Based on observation, a trading strategy is created, and then it is important to follow it.
This is fascinating to me, in particular,
"Overfitting.
Neural networks can develop a black box that will match past data perfectly; or with 100% probability as you like to say. They can do this with past lotto data. You can guess the results as the algo is walked forward in real-time."
Noob question here is
At what rate does the predictive value of such a thing decay?
This Overfitted application?
eg.
If the population of past data is statistically representative
and the past data analyzed stops drop dead on date time x,
at what rate does the predictive value of the application
decay from date time x, forward?
does the predictive value taper off starting at x+1 or
does it drop dead have No Predictive Value at x + 1?
At what rate does the predictive value of such a thing decay?
Actually, "probability" was someone else's choice word. I used it when I responded to their topics.
If the model is truly predictive, and there is no 'concept drift,' then there should be no degradation.
If the model is purely just (over)fitted, then there should be instant degradation.
If the model is truly predictive, but there is concept/model drift, then there will be degradation. The rate of degradation depends on whatever factors cause the drift, known and unknown; the calculus of those factors; and how well and often the system retrains/adapts the model to the drift, if at all.
Last edited:
To predict, you don't need to be 100% accurate.
If only 100% accuracy is called a prediction, weather forecasting is not a forecasting, because it is not 100% accurate.
Prediction is to get a high probability on the outcome of your action.
So your action is based on your assessment of probability on outcome.
Not based on randomness. When you react, you react based on higher probability, not on randomness.
Last edited:
I have come to understand that predicting is not that easy.
There are going to be charts and you need to read between the lines so to speak, so as to understand what seems to be going on. True you will try to think what will happen in the near future, but you can never be certain and so it will basically be a reaction based on intuition and your past experience, which will also be affected by your character, such as how willing you are to take a certain amount of risk.
There are going to be charts and you need to read between the lines so to speak, so as to understand what seems to be going on. True you will try to think what will happen in the near future, but you can never be certain and so it will basically be a reaction based on intuition and your past experience, which will also be affected by your character, such as how willing you are to take a certain amount of risk.
Because prediction is only part of a story. You can not win without prediction. But prediction alone would not make you winning.
It is math expectation that makes you winning.
Let's say you have a high probability of 80% on your trades. So for 10 trades you win 8 times and lose 2 times . But your average winning is $100 and average loss is $500. You are predicting well but still losing.
Your entries and exits decide average winning and loss. If you can not choose good entries and exits, even you have high probability on your trades, you could still lose.
Last edited: