Prediction intervals for ARIMA and structural time series models using importance sampling approach with uninformative priors for model parameters, leading to more accurate coverage probabilities in frequentist sense. Instead of sampling the future observations and hidden states of the state space representation of the model, only model parameters are sampled, and the method is based solving the equations corresponding to the conditional coverage probability of the prediction intervals. This makes method relatively fast compared to for example MCMC methods, and standard errors of prediction limits can also be computed straightforwardly.
|Date of publication||2016-03-17 13:14:01|
|Maintainer||Jouni Helske <firstname.lastname@example.org>|
acv_arma: Compute a theoretical autocovariance function of ARMA process
arima_pi: Prediction Intervals for ARIMA Processes with Exogenous...
avg_coverage_arima: Compute the average coverage of the prediction intervals...
avg_coverage_struct: Compute the average coverage of the prediction intervals...
dacv_arma: Compute the partial derivatives of theoretical autocovariance...
information_arma: Large Sample Approximation of Information Matrix for ARMA...
priors: Compute different types of importance weights based on...
struct_pi: Prediction Intervals for Structural Time Series with...
tsPI: Improved Prediction Intervals for ARIMA Processes and...
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.