Time Series Forecast [TSF]

Return to Complete List of Functions

The Time Series Forecast function predicts a future value of a field using the value of a given period linear regression line. Parameters ------------------ Data          The data to use in the regression. This is typically a field in a data series or a calculated value. Period        The number of bars of data to include in the regression, including the current value.                   For example, a period of 3 includes the current value and the two previous values. Offset        The number of bars into the future to predict.                   For example, an offset of 1 would predict the next value. Function Value ------------------------ The Time Series Forecast calculates probable future values of a data field by fitting a linear regression line over a given number of known values and following that line forward into the future. A linear regression line is a straight line which is as close to all of the given values as possible. Note that an offset of zero would return the current value on the given linear regression line. This is the same as the Linear Regression Indicator and the Time Series Moving Average. Also note that this type of prediction is purely mathematical. It does not involve neural networks or a training phase. It is simply the equivalent of drawing a line through the recent points and following that line forward. The Time Series Forecast at the beginning of a data series is not defined until there are enough values to fill the given period. Usage ---------- The Time Series Forecast can be used for predicting future values of fields. Since it represents the future value of fields, it can be plugged into indicators which take these fields to predict the future value of these indicators. Source ------------ This indicator is based on an entry in "Technical Analysis From A To Z" by Steven B. Achelis.