Journal of University of Chinese Academy of Sciences >
The Convergence Rates for Empirical Bayes Test of the Parameter in Scale Exponential Family
Department of Statistics and Finance, USTC
Received date: 1900-01-01
Revised date: 1900-01-01
Online published: 2007-01-15
Ｉｎ ｔｈｉｓ ｐａｐｅｒ，ｔｈｅ ｅｍｐｉｒｉｃａ ｌＢａｙｅｓ（ＥＢ） ｔｅｓｔ ｐｒｏｂｌｅｍ ｏｆ ｔｈｅ ｐａｒａｍｅｔｅｒ ｆｏｒｓｃａｌｅ ｅｘｐｏｎｅｎｔｉａｌ ｆａｍｉｌｙ ｕｎｄｅｒ ｗｅｉｇｈｅｄ ｌｉｎｅａｒ ｌｏｓｓ ｆｕｎｃｔｉｏｎｓｉｓ ｄｉｓｃｕｓｓｅｄ．Ｔｈｅ ＥＢ ｔｅｓ ｔｒｕｌｅｓ ｉｓｃｏｎｓｔｒｕｃｔｅｄ ｂｙ ｔｈｅ ｋｅｒｎｅｌ ｅｓｔｉｍａｔｉｏｎ ｍｅｔｈｏｄ ｂａｓｅｄ ｏｎ
Ｂｅｓｓｅｌ ｆｕｎｃｔｉｏｎｓ．Ｕｎｄｅｒｓｕｉｔａｂｌｅ ｃｏｎｄｉｔｉｏｎｓｔｈｅ ｐｒｏｐｏｓｅｄ ＥＢ ｔｅｓｔ ｒｕｌｅｓ ａｒｅ ａ ｓｙｍｐｔｏｔｉｃａｌｌｙ ｏｐｔｉｍａｌ ｗｉｔｈ ｔｈｅ ｃｏｎｖｅｒｇｅｎｃｅ ｒａｔｅ Ｏ（ｎ－１（ｌｎｎ）６）．Ｆｉｎａｌｌｙ，ａｎ ｅｘａｍｐｌｅ ｓａｔｉｓｆｙｉｎｇ ｔｈｅ ｃｏｎｄｉｔｉｏｎｓ ｏｆ ｔｈｅ ｔｈｅｏｒｅｍ ｉｓ ｓｈｏｗｎ．
WEI Li , WEI Lai-Sheng , KONG Sheng-Chun . The Convergence Rates for Empirical Bayes Test of the Parameter in Scale Exponential Family[J]. Journal of University of Chinese Academy of Sciences, 2007 , 24(1) : 9 -17 . DOI: 10.7523/j.issn.2095-6134.2007.1.002
/
| 〈 |
|
〉 |