TD_GENSERIES4FORMULA allows you to define and apply a formula to generate a time series. The formula is a mathematic expression that depends on a time index variable. The function takes the formula, and the length of the time series to generate. It calculates the value of the formula for each time index from 1 to the length of the time series, and returns a vector or array of the values.
You can use TD_GENSERIES4FORMULA to generate a simulated time series based on a model you have developed. You then use the simulated time series to make predictions about future values of the actual time series. It can also be used to create synthetic data for testing statistical models or machine learning algorithms. By generating time series with specific properties, such as non-stationary or correlated, you can test the performance of your models or algorithms under different conditions.
- Determine that the series to be modeled includes a trend.
- Use TD_LINEAR_REGR to curve fit the trend.
- Use the coefficients computed by TD_LINEAR_TREND, assemble a FORMULA and use TD_GENSERIES4FORMULA to generate a trend series.
- Use TD_BINARYSERIESOP to subtract the generated trend from the original series.