Linear Regression: Suitable for Forex Data? Difference between MQL5 Objects OBJ_REGRESSION and OBJ_STDDEVCHANNEL
1) Is linear regression actually suitable for Forex data? One assumption is that data must be randomly normally distributed and trending data isn't really. So, I assume, one have to transform the data first (detrending) and then apply linear Regression?
2) What is the difference between OBJ_REGRESSION and OBJ_STDEVCHANNEL? I implemented my own linear Regression and both objects create the same Regression line. However, Regression seems to have the channel based on 2 Standard devations whereas STDDEVCHANNEL is based on 1 Standard devivation.
1) That is totally in the realm of how the trader wishes to analyse the price data. For some it works, and for others not. It is very subjective.
2) They are similar in that both use a Linear Regression mean line of the equation "y = m*x + c", using Least Squares fitting. The difference is in the channel width, where the Linear regression channel is fixed and based on the extremes of the close price deltas (i.e. the maximum deviation of the Least Squares), while the Std.Dev. channel uses the mean deviation of the Least Squares, and for which you can then apply a ratio to set the channel width.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Hello Community,
I have two Questions:
1) Is linear regression actually suitable for Forex data? One assumption is that data must be randomly normally distributed and trending data isn't really. So, I assume, one have to transform the data first (detrending) and then apply linear Regression?
2) What is the difference between OBJ_REGRESSION and OBJ_STDEVCHANNEL? I implemented my own linear Regression and both objects create the same Regression line. However, Regression seems to have the channel based on 2 Standard devations whereas STDDEVCHANNEL is based on 1 Standard devivation.