| On the correlation of nonlinear variables containing secular trend variations： numerical experiments |
| |
| 引用本文： | 施能,易燕明,顾骏强,夏冬冬. On the correlation of nonlinear variables containing secular trend variations： numerical experiments[J]. 中国物理, 2006, 15(9): 2180-2184. DOI: 10.1088/1009-1963/15/9/046 |
| |
| 作者姓名： | 施能 易燕明 顾骏强 夏冬冬 |
| |
| 作者单位： | (1)Guangzhou Institute of Tropical and Oceanic Meteorology,Guangzhou 510080, China; (2)Key Laboratory of Meteorology and Environment, Nanjing University of Information Science and Technology, Nanjing 210044, China; (3)National Marine Environment Forecast Center, Beijing 100081, China; (4)Zhejiang Meteorological Science Research Institute, Hanzhou 310017, China |
| |
| 基金项目： | Project supported by the NationalNatural Sciences Foundation of China (Grant Nos 40275031 and90411008). |
| |
| 摘 要： | Due to global warming, the general circulation, underlying surfaces characteristics,and geophysical and meteorological elements all show evident secular trends. Thispaper points out that when calculating the correlation of two variables containingtheir own obvious secular trends, the interannual correlation characteristicsbetween the two variables may be distorted (overestimated or underestimated).Numerical experiments in this paper show that if two variables have opposite seculartrends, the correlation coefficient between the two variables is reduced (thepositive correlation is underestimated, or the negative correlation isoverestimated); and if the two variables have the same sign of secular trends, thecorrelation coefficient between the two variables is increased (the positivecorrelation is overestimated, or the negative correlation is underestimated).Numerical experiments also suggest that the effect of secular trends on theinterannual correlation of the two variables is interchangeable, that is to say, aslong as the values of the two trends are not changed, the two variables interchangetheir positions, and the effect of the secular trends on the interannual correlationcoefficient of the two variables remains the same. If the two variables have thesame-(opposite-) sign trends, the effect of secular trends on the interannualcorrelation coefficient is more (less) distinctive. A meteorological example isgiven.
|
| 关 键 词： | 相关性分析 长期倾向 倾向系数 数字实验 |
| 收稿时间： | 2006-03-21 |
| 修稿时间： | 2006-03-212006-05-29 |
On the correlation of nonlinear variables containing secular trend variations: numerical experiments |
| |
| Shi Neng,Yi Yan-Ming,Gu Jun-Qiang and Xia Dong-Dong. On the correlation of nonlinear variables containing secular trend variations: numerical experiments[J]. Chinese Physics, 2006, 15(9): 2180-2184. DOI: 10.1088/1009-1963/15/9/046 |
| |
| Authors: | Shi Neng Yi Yan-Ming Gu Jun-Qiang Xia Dong-Dong |
| |
| Affiliation: | Key Laboratory of Meteorology and Environment, Nanjing University of Information Science and Technology, Nanjing 210044, China; Guangzhou Institute of Tropical and Oceanic Meteorology,Guangzhou 510080, China; Zhejiang Meteorological Science Research Institute, Hanzhou 310017, China; National Marine Environment Forecast Center, Beijing 100081, China |
| |
| Abstract: | Due to global warming, the general circulation, underlying surfaces characteristics,and geophysical and meteorological elements all show evident secular trends. Thispaper points out that when calculating the correlation of two variables containingtheir own obvious secular trends, the interannual correlation characteristicsbetween the two variables may be distorted (overestimated or underestimated).Numerical experiments in this paper show that if two variables have opposite seculartrends, the correlation coefficient between the two variables is reduced (thepositive correlation is underestimated, or the negative correlation isoverestimated); and if the two variables have the same sign of secular trends, thecorrelation coefficient between the two variables is increased (the positivecorrelation is overestimated, or the negative correlation is underestimated).Numerical experiments also suggest that the effect of secular trends on theinterannual correlation of the two variables is interchangeable, that is to say, aslong as the values of the two trends are not changed, the two variables interchangetheir positions, and the effect of the secular trends on the interannual correlationcoefficient of the two variables remains the same. If the two variables have thesame-(opposite-) sign trends, the effect of secular trends on the interannualcorrelation coefficient is more (less) distinctive. A meteorological example isgiven. |
| |
| Keywords: | correlation analysis secular trend trend coefficient numerical experiment |
| 本文献已被 维普 等数据库收录！ |
| 点击此处可从《中国物理》浏览原始摘要信息 |
|
点击此处可从《中国物理》下载免费的PDF全文 |
|
