int ARMAMLE(TVec *Data, TVec *P, TVec *T, TVec *Residuals, double &MLE, double &mu);
Estimate ARMA process AR and MA coefficients.
| # | Name | Type | Description |
|---|---|---|---|
| 1 | Data | TVec * | Time series data set. |
| 2 | P | TVec * | ARIMA Before call stores initial estimates for ARIMA Phi coefficients. After call returns MLE estimates for Phi coefficients without leading 1.0. |
| 3 | T | TVec * | ARIMA Before call stores initial estimates for ARIMA Theta coefficients. After call returns MLE estimates for Theta coefficients. |
| 4 | Residuals | TVec * | Returns residuals between predicted (MLE) and actual time series values. |
| 5 | MLE | double & | Returns -2 log likelihood of ARMA model. |
| 6 | mu | double & | Returns the estimated modified series average value (constant). |
Returns: Number of evaluations needed to converge to MLE solution.
Remarks:
Estimate ARMA(p,t) process coefficients by using MLE.
Declared in Dew::Stats::Units::StatTimeSerAnalysis · Dew.Stats/Units.StatTimeSerAnalysis.h · Cross-compiler