日本語
Español
中文
LINEAR REGRESSION
Doug Schaff
FX-Strategy
| Characteristic: | Trend Direction Indicator | ||
| Parameter Defaults: | LR Period | 9 | controls the measurement period for the regressionl |
| Plots: | LR | ||
Linear Regression is a statistical technique that uses the least squares method to fit a straight line through scattered data points. By taking these points in a dynamic manner on each price bar, a line similar to a moving average is created and plotted.
The Linear Regression Line indicator was developed to statistically help determine where a market's price might be in the near future using current and past price history. Some users believe that when prices rise above or fall below the linear regression line, they are overextended and will begin to move back towards the line. It may therefore be used to suggest that a retracement in a directional price action may occur.
If prices trend upwards, linear regression attempts to statistically calculate what the upward bias of the price may be relative to the current price. If prices trends downwards, it will attempt to statistically calculate the downward bias of the price.
On a chart a Linear Regression line is basically the same as a moving average but it will be noted that it reacts very quickly to changes in price direction and holds closer to price action displaying the tendency for prices to move back towards the regression line if there is any larger move away.
While Linear Regression may be used in the same way as a moving average it must be understood that the speed of reversal can cause whipsaws to occur.
