Linear regression may sound intimidating, but the mathematical concept is a
simple one. All this technique does is fit a straight line through a finite
number of data points by minimizing the sum of the squared vertical distance
between the line and each of the points. In our context, this means that if
time is represented by days on the horizontal axis and the closing price
on those days is plotted as dots on the vertical axis (a normal closing price
chart), then we try to fit a straight line through those closing-price dots
such that the total sum of the squared vertical distance between each closing
price and the line are minimized. This would then be our best-fit line.
Raff regression channel Raff Regression Channels show the range prices can be expected to deviate from a Linear Regression trend line. Developed by Gilbert Raff, the regression channel is a line study the plots directly on the price chart. The Regression Channel provides a precise quantitative way to define a price trend and its boundaries. The Regression Channel is constructed by plotting two parallel, equidistant lines above and below a Linear Regression trend line.
The distance between the channel lines to the regression line is the greatest distance that any one high or low price is from the regression line.
Raff Regression Channels contain price movement, with the bottom channel line providing support and the top channel line providing resistance. Prices may extend outside of the channel for a short period of time. However, if prices remain outside the channel for a long period of time, a reversal in trend may be imminent.