<i>K</i>-NEAREST NEIGHBOR NONPARAMETRIC REGRESSION FOR PROBABILITY FORECASTING WITH ITS APPLICATIONS

Zhai Yumei; Zhao Ruixing; Xiao Renchun; Wang Liwei

Vol.16, NO.4, 2005

Article Contents

Article Navigation > Journal of Applied Meteorological Science > 2005 > 16(4): 453-460

Zhai Yumei, Zhao Ruixing, Xiao Renchun, et al. K-nearest neighbor nonparametric regression for probability forecasting with its applications. J Appl Meteor Sci, 2005, 16(4): 453-460.

Citation:

Zhai Yumei, Zhao Ruixing, Xiao Renchun, et al. K-nearest neighbor nonparametric regression for probability forecasting with its applications. J Appl Meteor Sci, 2005, 16(4): 453-460.

Zhai Yumei, Zhao Ruixing, Xiao Renchun, et al. K-nearest neighbor nonparametric regression for probability forecasting with its applications. J Appl Meteor Sci, 2005, 16(4): 453-460.

Citation:

Zhai Yumei, Zhao Ruixing, Xiao Renchun, et al. K-nearest neighbor nonparametric regression for probability forecasting with its applications. J Appl Meteor Sci, 2005, 16(4): 453-460.

PDF( 156 KB)

K-NEAREST NEIGHBOR NONPARAMETRIC REGRESSION FOR PROBABILITY FORECASTING WITH ITS APPLICATIONS

Beijing Institute of Applied Meteorology , Beijing 100029

Received Date: 2004-07-26
Rev Recd Date: 2005-05-23
Publish Date: 2005-08-31

Abstract

Abstract

Although probability forecasts based on a parametric regression scheme have good fitting rates the results are not so stable. For this reason, a new approach is proposed to such forecasts by means of a K-nearest neighbor nonparametric regression technique, and the technique includes 4 main components such as a database of historical samples, production of nearest neighbor subsets, their optimization and estimate of predictands. Case experiments are conducted on univariate (cloudiness or precipitation) and multivariate joint (e. g., rainfall, total cloudiness, wind speed and temperature) probability forecasting, with the results tested. Results show that forecasts from the nonparametric regression scheme are high-stability, with good prospects in operational weather forecast.
- Analogue forecasting;
- Nearest neighbor;
- Nonparametric regression estimation;
- Probability forecasting

FullText(HTML)

References(8)

Figures and Tables

图 1 非参数估计算法示意图

DownLoad: Full-Size Img PowerPoint

图 2 1996 年北京地区降水有无预报准确率

DownLoad: Full-Size Img PowerPoint

图 3 1996 年北京地区分级降水、总云量和低云量的有序概率得分

DownLoad: Full-Size Img PowerPoint

图 4 日降水量的有序概率得分

DownLoad: Full-Size Img PowerPoint

表 1 K 近邻非参数回归预报技术与相似预报技术的预报结果比较

表 2 2002 年7 月北京站“ 好天”概率预报评价结果

References

[1]	Schaal Stefan. Nonparametric Regression for Learning. Proceeding of the Conference on Prerational Intelligence. Germany, 1994.
[2]	Altman N S. An introduction to kernel and nearest neighbor nonparametric regression. The American Statistician, 1992, 46(3):175-185. https://www.researchgate.net/publication/238181607_An_Introduction_to_Kernel_and_Nearest-Neighbor_Nonparametric_Regression
[3]	Oswald R K, William T S, Brian L S. Traffic Flow Forecasting Using Approximate Nearest Neighbor Nonparametric Regression. Research Report, No. UVACTS-15-13-7, 2001.
[4]	施能. 气象科研与预报中的多元分析方法. 北京:气象出版社,1995. 85-88.
[5]	Yianilos P N. Data Structure and Algorithms for Nearest Neighbor Search in General Metric Space. Proceeding of the ACM-SIAM Symposium on Discrete Algorithms. 1993. 311-321. https://www.researchgate.net/publication/2596667_Data_Structures_and_Algorithms_for_Nearest_Neighbor_Search_in_General_Metric_Spaces
[6]	罗阳.一种新的相似性度量--高分辨相似系数.空军气象学院学报,1996,17(1):23-32.
[7]	朱盛明,曲学实. 数值预报产品统计解释技术的进展. 北京:气象出版社,1988. 163-165.
[8]	张军,葛军,田俊杰,等. 概率天气预报及其应用. 北京:气象出版社,1998. 29-43.