Forecasting Global Temperature Variations by Neural Networks
Global temperature variations between 1861 and 1984 are forecast usingsregularization networks, multilayer perceptrons and linearsautoregression. The regularization network, optimized by stochasticsgradient descent associated with colored noise, gives the bestsforecasts. For all the models, predicti...
| Main Authors: | , |
|---|---|
| Language: | en_US |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/7208 |
| _version_ | 1826211346425839616 |
|---|---|
| author | Miyano, Takaya Girosi, Federico |
| author_facet | Miyano, Takaya Girosi, Federico |
| author_sort | Miyano, Takaya |
| collection | MIT |
| description | Global temperature variations between 1861 and 1984 are forecast usingsregularization networks, multilayer perceptrons and linearsautoregression. The regularization network, optimized by stochasticsgradient descent associated with colored noise, gives the bestsforecasts. For all the models, prediction errors noticeably increasesafter 1965. These results are consistent with the hypothesis that thesclimate dynamics is characterized by low-dimensional chaos and thatsthe it may have changed at some point after 1965, which is alsosconsistent with the recent idea of climate change.s |
| first_indexed | 2024-09-23T15:04:28Z |
| id | mit-1721.1/7208 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T15:04:28Z |
| publishDate | 2004 |
| record_format | dspace |
| spelling | mit-1721.1/72082019-04-10T11:52:45Z Forecasting Global Temperature Variations by Neural Networks Miyano, Takaya Girosi, Federico time series prediction chaotic systems neural nets RBF Global temperature variations between 1861 and 1984 are forecast usingsregularization networks, multilayer perceptrons and linearsautoregression. The regularization network, optimized by stochasticsgradient descent associated with colored noise, gives the bestsforecasts. For all the models, prediction errors noticeably increasesafter 1965. These results are consistent with the hypothesis that thesclimate dynamics is characterized by low-dimensional chaos and thatsthe it may have changed at some point after 1965, which is alsosconsistent with the recent idea of climate change.s 2004-10-20T20:49:51Z 2004-10-20T20:49:51Z 1994-08-01 AIM-1447 CBCL-101 http://hdl.handle.net/1721.1/7208 en_US AIM-1447 CBCL-101 11 p. 342101 bytes 403018 bytes application/octet-stream application/pdf application/octet-stream application/pdf |
| spellingShingle | time series prediction chaotic systems neural nets RBF Miyano, Takaya Girosi, Federico Forecasting Global Temperature Variations by Neural Networks |
| title | Forecasting Global Temperature Variations by Neural Networks |
| title_full | Forecasting Global Temperature Variations by Neural Networks |
| title_fullStr | Forecasting Global Temperature Variations by Neural Networks |
| title_full_unstemmed | Forecasting Global Temperature Variations by Neural Networks |
| title_short | Forecasting Global Temperature Variations by Neural Networks |
| title_sort | forecasting global temperature variations by neural networks |
| topic | time series prediction chaotic systems neural nets RBF |
| url | http://hdl.handle.net/1721.1/7208 |
| work_keys_str_mv | AT miyanotakaya forecastingglobaltemperaturevariationsbyneuralnetworks AT girosifederico forecastingglobaltemperaturevariationsbyneuralnetworks |