We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Predicting realizations of daily weather data for climate forecasts using the non-parametric nearest-neighbour re-sampling technique.
- Authors
Bannayan, Mohammad; Hoogenboom, Gerrit
- Abstract
The article presents a study that verifies the k-nearest neighbours (k-NN) approach for the prediction of daily weather consequences. The method can be implemented on the assumption that the weather during the target year is similar to the weather recorded in the past. It uses the nearest-neighbour re-sampling method for the simultaneous forecast of daily radiation, maximum and minimum temperature, and precipitation for multiple locations. The study concluded that the k-NN approach on the basis of pattern recognition can be considered as a reliable method to forecast daily weather sequences according historical weather data.
- Subjects
WEATHER forecasting; PRECIPITATION forecasting; TEMPERATURE; METHODOLOGY; DATA; RECORDS; METEOROLOGY; SAMPLING (Process); RESEARCH
- Publication
International Journal of Climatology, 2008, Vol 28, Issue 10, p1357
- ISSN
0899-8418
- Publication type
Article
- DOI
10.1002/joc.1637