Home > Search Results: keywords:"short-term forecasting"
Search:
Search Tips :: Advanced Search
Search collections:
Sort by: Display results: Output format:
National Repository of Grey Literature 2 records found  Search took 0.00 seconds. 
Short-term Electric Load Forecasting Using Czech Data
 Řanda, Martin ; Krištoufek, Ladislav (advisor) ; Čech, František (referee)
Forecasting electric load accurately is a critical prerequisite to dependable power grid operation. It is thus in the best interests of the responsible institutions to develop and maintain performant models for predicting load. In this thesis, we analyze Czech electric load data and execute three pseudo-out-of-sample forecasting exercises. We employ standard econometric as well as machine learning methods and compare the results to benchmarks, including the predictions published by the Czech transmission system operator. The results of the first task examining the predictability of minute loads using 11 years of data indicate that the high-frequency load series is predictable. In the second and third exercises, we utilize hourly loads with additional explanatory variables. We generate one-step-ahead and 48-hours-ahead forecasts on the 2021 out- of-sample set and evaluate the performance of several methods. In both exercises, the most accurate results are produced by averaging forecasts of our specified recurrent neural network and the seasonal autoregressive integrated moving average model, achieving a mean absolute percentage error of less than 0.5% on the out-of-sample set in the one-step-ahead analysis and 2.3% in the 48-hours-ahead exercise, outperforming the operator's predictions.
Detailed record
Forecasting Czech GDP Using Mixed-Frequency Data Models
 Franta, Michal ; Havrlant, David ; Rusnák, Marek
In this paper we use a battery of various mixed-frequency data models to forecast Czech GDP. The models employed are mixed-frequency vector autoregressions, mixed-data sampling models, and the dynamic factor model. Using a dataset of historical vintages of unrevised macroeconomic and financial data, we evaluate the performance of these models over the 2005–2012 period and compare them with the Czech National Bank’s macroeconomic forecasts. The results suggest that for shorter forecasting horizons the accuracy of the dynamic factor model is comparable to the CNB forecasts. At longer horizons, mixed-frequency vector autoregressions are able to perform similarly or slightly better than the CNB forecasts. Furthermore, moving away from point forecasts, we also explore the potential of density forecasts from Bayesian mixed-frequency vector autoregressions.
Fulltext: Download fulltextPDF
Detailed record

Interested in being notified about new results for this query?
Subscribe to the RSS feed.
Digital Repository :: Search :: Submit :: Personalize :: Help
Powered by Invenio v1.1.2
Maintained by admin.invenio@techlib.cz
This site is also available in the following languages:
Česky  English