Indicators: Center of Gravity J. F. Ehlers - page 2
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
I haven’t read Ehlers, but it seems I haven’t missed out on anything. It’s all so convoluted and incomprehensible that I’ll try to put his ideas into terms that a person can understand.
So, the formula for any linear indicator is: sum(c[i]*price[i])/sum(c[i]). That’s all correct and there are no issues there. It’s worth mentioning, however, that there is one important restriction. For all the indicator’s coefficients, the condition -1 < c[i] /sum(c[i]) < 1 must hold– only then will the indicator be stable. Furthermore, you can check the indicator for accuracy. Take the smallest coefficient (if there are negative coefficients, consider their absolute value; if there are coefficients equal to zero, disregard them). The inequality 100,000*cmin ≤sum(c[i]) must hold. If this condition is not met, the indicator’s period is excessive and can be reduced without any loss of accuracy.
Next, we calculate the indicator’s centre of gravity, if necessary. This can be found using the formula: sum(c[i]*(i+1))/sum(c[i]), i=0...period-1. The smaller the resulting value, the faster the indicator reacts to price changes, and the lower its lag. The resulting centre of gravity is always static and does not change.
Now to Ehlers’s idea itself. A conventional indicator operates in the price domain. Ehlers decided to examine how the indicator behaves in the time domain. We rewrite the indicator’s formula as follows: sum(price[i]*t[i])/sum(price[i]), where t[i] = period - i. As a reference (zero) point, we can take the indicator’s centre of gravity (in this case, 0.5 × period, but if we use coefficients, we can obtain other interesting variations), and plot the resulting value relative to it. In this case, if the indicator is above this point, the trend is up; below it, the trend is down; and around zero, the market is flat. You can add normalisation by the indicator’s period, which allows you to compare several indicators with different periods within a single window.