Charts are tangible, objective and predictable! Test me!
- Thread starter rajesheck
- Start date
- Status
@rajesheck
If the first hidden chart was partial and disqualified for reading, you should have flatly declined to read it.
If what you wrote about the second hidden chart is true, you should have DRAWN the second support/resistance breakout levels. (Since I do not trade options, I have no idea what would have happened had a trader remained in this position TOO long.)
If what you wrote about the third one was true, you should have DRAWN the second support breakout level, and DRAWN the third support breakout level.
All this talk AFTER the fact, from my perspective, is just so much blather. You might be a serious trader, but to me, this thread kind of feels more like a parlor game.
If the first hidden chart was partial and disqualified for reading, you should have flatly declined to read it.
If what you wrote about the second hidden chart is true, you should have DRAWN the second support/resistance breakout levels. (Since I do not trade options, I have no idea what would have happened had a trader remained in this position TOO long.)
If what you wrote about the third one was true, you should have DRAWN the second support breakout level, and DRAWN the third support breakout level.
All this talk AFTER the fact, from my perspective, is just so much blather. You might be a serious trader, but to me, this thread kind of feels more like a parlor game.
Maybe you meant to say "No one can predict the future with 100 % accuracy 100 % of the time". If so, I obviously agree.
Prediction/forecasting is generally about probabilities, though. Even if you have a high probability prediction at 80 % chance, you'll still be wrong 2 times out of 10.
It seems to be a controversial topic on ET, but even a basic trend system is based on prediction in the sense that you're expecting the trend to continue.
And for those using set-ups (tested and trialed, I assume), there's still a probability set associated with them, i.e., it's successful 6 out of 10 times in back-testing/forward-testing with a 2:1 win ratio. Essentially, you're predicting an outcome with known probabilities...
I like prediction.
If you have noticed i responded to one of the post about devil's details...i said my predictions are based on second support/resistance breakout.
My style of communication is not to flood overload info and instead build conversation as needed by the group.
To create a more advanced predictive indicator in MultiCharts PowerLanguage, you can utilize machine learning techniques such as regression or classification algorithms. Here's an example of how you can implement a linear regression-based predictive indicator:
powerlanguage
inputs:
LookbackPeriod(10); // Number of bars used for training
variables:
RegressionCoeff(0), // Coefficient of the regression line
Intercept(0), // Intercept of the regression line
Prediction(0); // Predicted value
if CurrentBar > LookbackPeriod then begin
// Perform linear regression using the past data
LinearRegression(Close, LookbackPeriod, RegressionCoeff, Intercept);
// Predict the next bar's value
Prediction = RegressionCoeff * (CurrentBar - LookbackPeriod) + Intercept;
// Plot the regression line
Plot1(RegressionCoeff * (CurrentBar - LookbackPeriod) + Intercept, "Regression Line", blue);
// Plot the predicted value
Plot2(Prediction, "Prediction", green);
// Generate buy or sell signals based on the predicted value
if Close > Prediction then
Buy("BuySignal") next bar at market
else if Close < Prediction then
SellShort("SellSignal") next bar at market;
end;
In this example, the indicator uses the LinearRegression function to perform linear regression on the past LookbackPeriod bars of the closing prices. It then calculates the predicted value for the next bar based on the regression line equation. The indicator plots both the regression line and the predicted value on the chart. It also generates buy or sell signals based on whether the current price is above or below the predicted value.
Keep in mind that this is a simplified example, and you may need to adapt it based on your specific requirements and the machine learning techniques you intend to use. Additionally, you may need to consider other factors such as data preprocessing, feature selection, and model evaluation to build a more robust and accurate predictive indicator.
powerlanguage
inputs:
LookbackPeriod(10); // Number of bars used for training
variables:
RegressionCoeff(0), // Coefficient of the regression line
Intercept(0), // Intercept of the regression line
Prediction(0); // Predicted value
if CurrentBar > LookbackPeriod then begin
// Perform linear regression using the past data
LinearRegression(Close, LookbackPeriod, RegressionCoeff, Intercept);
// Predict the next bar's value
Prediction = RegressionCoeff * (CurrentBar - LookbackPeriod) + Intercept;
// Plot the regression line
Plot1(RegressionCoeff * (CurrentBar - LookbackPeriod) + Intercept, "Regression Line", blue);
// Plot the predicted value
Plot2(Prediction, "Prediction", green);
// Generate buy or sell signals based on the predicted value
if Close > Prediction then
Buy("BuySignal") next bar at market
else if Close < Prediction then
SellShort("SellSignal") next bar at market;
end;
In this example, the indicator uses the LinearRegression function to perform linear regression on the past LookbackPeriod bars of the closing prices. It then calculates the predicted value for the next bar based on the regression line equation. The indicator plots both the regression line and the predicted value on the chart. It also generates buy or sell signals based on whether the current price is above or below the predicted value.
Keep in mind that this is a simplified example, and you may need to adapt it based on your specific requirements and the machine learning techniques you intend to use. Additionally, you may need to consider other factors such as data preprocessing, feature selection, and model evaluation to build a more robust and accurate predictive indicator.
I realize our time is valuable here on ETMaybe you meant to say "No one can predict the future with 100 % accuracy 100 % of the time". If so, I obviously agree.
Prediction/forecasting is generally about probabilities, though. Even if you have a high probability prediction at 80 % chance, you'll still be wrong 2 times out of 10.
It seems to be a controversial topic on ET, but even a basic trend system is based on prediction in the sense that you're expecting the trend to continue.
And for those using set-ups (tested and trialed, I assume), there's still a probability set associated with them, i.e., it's successful 6 out of 10 times in back-testing/forward-testing with a 2:1 win ratio. Essentially, you're predicting an outcome with known probabilities...
I like prediction.
but my comment went on beyond the first sentence.No problem there. More like underflow info.
May the group that needs your conversation be greatly edified by the contribution of your information-light communication style. Clearly, I visited the wrong thread (I thought the whole point was to show that charts are tangible, objective and predictable) but more power to you, my good man.
- Status