BEKK GARCH output
Estimates Std.Error t value Pr{>|t|}
B12 0.90 NaN NaN NaN
A NaN (Not a Number) result in a BEKK GARCH model in R may indicate one of the following problems:
1_Missing or incorrect data: Make sure that your data is complete and correct. If there are missing values in your data, you may need to either impute the missing values or exclude the observations with missing values
2-Convergence issues: BEKK GARCH models can be sensitive to the starting values of the parameters. You can try different starting values and see if this resolves the issue. Alternatively, you can use a different optimization algorithm, such as the BFGS or Nelder-Mead method, which are more robust to convergence issues
3-Non-stationary data: The BEKK GARCH model assumes that the data is stationary. If your data is non-stationary, you may need to first difference the data or use a different model, such as the EGARCH or APARCH model
4-Specification errors: The BEKK GARCH model has several assumptions and specifications, such as the structure of the conditional covariance matrix and the distribution of the errors. Make sure that your model specification is correct and meets the necessary assumptions
If none of these steps resolve the issue, you may need to consult with a specialist in econometrics or financial time series analysis to troubleshoot further.
- Badges
- Science topic
- Similar topics
- Academic Journals