WEKO3
アイテム
競馬データにみられる統計的偏りについて （2）
http://hdl.handle.net/10271/17
http://hdl.handle.net/10271/17aca57a5f-302d-4dc2-9442-322f073cf906
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
kiyo19_1.pdf (553.2 kB)
|
|
| Item type | 紀要論文 / Departmental Bulletin Paper(1) | |||||||
|---|---|---|---|---|---|---|---|---|
| 公開日 | 2013-08-27 | |||||||
| タイトル | ||||||||
| タイトル | 競馬データにみられる統計的偏りについて （2） | |||||||
| 言語 | ||||||||
| 言語 | jpn | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | adjacent interval of three numbers | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | chi-square test | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | contingency table | |||||||
| キーワード | ||||||||
| 主題Scheme | Other | |||||||
| 主題 | exchangeable random variables | |||||||
| 資源タイプ | ||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||||
| 資源タイプ | departmental bulletin paper | |||||||
| その他のタイトル | ||||||||
| その他のタイトル | On the Statistical Bias Found in the Horse Racing Data (2) | |||||||
| 著者 |
野田, 明男
× 野田, 明男
|
|||||||
| 書誌情報 |
浜松医科大学紀要. 一般教育 巻 19, p. 1-7, 発行日 2005-02-25 |
|||||||
| 出版者 | ||||||||
| 出版者 | 浜松医科大学 | |||||||
| 抄録 | ||||||||
| 内容記述タイプ | Abstract | |||||||
| 内容記述 | This is a continuation of the author’s previous paper [3]. Taking a new point of view based on exchangeable random variables ti (i =1, 2, 3, 4) defined below, we report what type of statistical bias can be found in the horse racing data [4]. In order to define these exchangeable random variables, let us consider a racing with m participants, and denote by {a, b, c}(1 ≦ a < b < c ≦ m) a set of the numbers of the horses finishing in the top three. We recall the null hypothesis H0 in [3] : This set {a,b,c} is a result of random sampling from the population {1, 2, …, m}. Now we define t1= a, t2 = b ? a, t3= c ? b, and t4 = m+1? c. Then the probability distribution of (ti1, ti2, ti3, ti4) under H0 turns out to be the same for any permutation σ = (i1, i2, i3, i4) of the set {1, 2, 3, 4}. By using the order statistics t(i), we form a suitable decomposition for the total event Ω consisting of mC3 simple events; for example, in the case of m =16 and m =15, we have 16C3 =11×4!+(12+7+4)× 4 2 ! +(4+1)× 4 3 ! ! and 15C3 =9×4!+(9+6+2)× 4 2 ! +3×4C2+(3+1)× 4 3 ! !+1 ×1, respectively. This decomposition and its implications are discussed in §1. In §2 we take up five racetracks, Chukyo, Hanshin, Kyoto, Nakayama and Tokyo, to examine all racings of m =16 carried out on these racetracks. Indeed, the above-mentioned decomposition leads us to sum up the original data [4] into 22 kinds of contingency tables for every racetrack. Performing the chisquare tests for all contingency tables thus obtained, we arrive at the following result on the P-values : | |||||||
| ISSN | ||||||||
| 収録物識別子タイプ | ISSN | |||||||
| 収録物識別子 | 09140174 | |||||||
| NII書誌ID | ||||||||
| 収録物識別子タイプ | NCID | |||||||
| 収録物識別子 | AN10032827 | |||||||
| フォーマット | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | application/pdf | |||||||
| 著者版フラグ | ||||||||
| 出版タイプ | VoR | |||||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||
