Machine learning in trading: theory, models, practice and algo-trading - page 3541
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
No.
there's no smoothing and there's no lag at all.
But we can say there is a filtering of trend noise
The picture shows
that on the original the price goes up with noise, and on the reconstructed series you can see that all candles are "white" and you can filter by this feature
property number 2
Even more interesting...
check out the hepa candles
All that is left is to come up with a trading system that will show expectation of at least more than 5 spreads and you can start browsing catalogues of yachts for sale.
no formulas, pure common sense and practice.....
You can't describe the high-frequency oscillations that it suppressed with a single flywheel, the size of the window won't help, it's irretrievably lost information....
there is not so much information lost as it seems :-) in absolute accuracy it is restored: knowing 1 price point + MA line, you can get a price line (MA-scheme is incremental, you can spin it back).
unit value of weighted MA = sum of vector prices * vector weights
for 2 periods of MA we get a square matrix (with coeff in ribbon form). It remains to solve the equation
there is not so much information lost as it seems :-) it is restored in absolute accuracy: knowing 1 price point + MA line,
I don't give OHLC model prices, so we don't know any price point.
I don't give prices for OHLC models, so we don't know any price points.
The MA calculation takes the price vector, multiplies it by the coefficient matrix, and gets the MA results.
you give it the vector of MA results, it knows the c const coefficient matrix (MA) and it solves the inverse equation :-)
PS/ in MA (SMA,LWMA) information dies only about errors and limits of double. When weights are unambiguous and there are no thresholds and zero coefficients, exactly
When calculating MA, the price vector is multiplied by the coefficient matrix and MA results are obtained.
you give her the vector of MA results, she knows the c const coefficient matrix (MA) and she solves the inverse equation :-)
PS/ in MA (SMA,LWMA) information dies only about errors and limits of double. When weights are unambiguous and there are no thresholds and zero coefficients, exactly
there is no need to invent something that is not there...
AMO only has access to one last value of the bag and the last value of RSI... ALL!!!!
there are no matrices, matrixes.
What does the new data show?
on the model, the model doesn't diverge, because it's a ruler.
on trading, I haven't formulated trading rules yet.
I have described gaps in some way (through if else), but this is just the beginning, I have not applied MO to trading rules yet.
If nothing else, it is an ordinary interpolation (approximation) with a given degree or accuracy. The degree or accuracy is set by the number of features.
The point is not that at all, but that from the price we can get a new price (price model) and it can be investigated.
it can be "cleaner/quality" than the original.
also act as a new source of features and trade rules.
+
there is no overfitting, the model without parameters is purely based on formulas.
+
during night hours when prices are "torn" the price model works better than the original one.
on the model, the model doesn't diverge, because it's a line.
on trading, I haven't formulated trading rules yet.
I have described the gaps in some way (through if else), but this is just the beginning, I have not applied the MO to the trading rules yet.
The point is not that at all, but that from the price we can get a new price (price model) and it can be investigated.
It can be "cleaner/quality" than the original.
It can also act as a new source of features and trading rules.
You cycle through these models and calculate the statistics of deviations from the source. For longs and shorts. This way you can pick up the TC, and then the yacht.
I don't get it, deviation of what from what?