國立高雄大學圖資館 |

Language: English

Back

條件常態分佈模型之相容性探討 = Compatibility of Co...

國立高雄大學統計學研究所

條件常態分佈模型之相容性探討 = Compatibility of Conditional Normal Distribution Model

Record Type:	Language materials, printed : monographic
Paralel Title:	Compatibility of Conditional Normal Distribution Model
Author:	曾婉芸,
Secondary Intellectual Responsibility:	國立高雄大學
Place of Publication:	[高雄市]
Published:	撰者;
Year of Publication:	2013[民102]
Description:	55面圖，表格 : 30公分;
Subject:	相容性
Subject:	Compatibility
Online resource:	http://handle.ncl.edu.tw/11296/ndltd/45849507622224395914
Notes:	參考書目：面26-27
Notes:	102年10月31日公開
Summary:	我們知道多維常態分佈的條件分佈皆是常態分佈，然而在給定條件分佈是常態的情形下，其相對應的聯合分佈不一定存在（此情形稱為不相容），即使存在（此情形稱為相容）也不一定是常態分佈。在本研究中，我們將探討條件常態分佈模型是相容的充要條件，同時，在相容的情形下，我們將刻劃所對應的聯合分佈之形式。另一方面，有研究者探討在給定的條件分佈模型是不相容的情形下，試圖計算一個聯合分佈，使得此聯合分佈之條件分佈能盡量地接近所給定的條件分佈；也有研究者提出以不相容的條件分配模型進行缺失資料插補，實證效果也有出人意外的表現。在本次研究中，我們將透過模擬的方式，探討上述兩結果。 We know that all conditionals of a multivariate normal distribution are normal distributions. However, if the given conditional distributions are normal, there may be no joint distribution having these conditionals (we call this case to be incompatible). Even though there exists a corresponding joint distribution (we call this case to be compatible), it may not be a normal distribution. In this studying, we will provide the necessary and sufficient conditions for compatibility of conditional normal distribution models. In addition, under compatibility, we will characteristize the corresponding joint distribution.On the other hand, when the given conditional distributions are incompatible, some researchers try to compute one joint distribution whose conditional distributions are as close as possible to the given conditional distributions. In addition, some researchers use incompatible conditional distributions to impute the missing data, and the empirical result is well unexpectedly. In this paper, we will discuss the above outcomes.

條件常態分佈模型之相容性探討 = Compatibility of Conditional Normal Distribution Model
曾, 婉芸

條件常態分佈模型之相容性探討 = Compatibility of Conditional Normal Distribution Model / 曾婉芸撰 - [高雄市] : 撰者, 2013[民102]. - 55面 ; 圖，表格 ; 30公分.
參考書目：面26-27102年10月31日公開.
相容性Compatibility

條件常態分佈模型之相容性探討 = Compatibility of Conditional Normal Distribution Model
LDR:03008nam0a2200289 450 001 389654
005 20170214095500.0
009 389654
010 0 $b 精裝
010 0 $b 平裝
100 $a 20170214d2013 k y0chiy05 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 條件常態分佈模型之相容性探討 $d Compatibility of Conditional Normal Distribution Model $z eng $f 曾婉芸撰
210 $a [高雄市] $c 撰者 $d 2013[民102]
215 0 $a 55面 $c 圖，表格 $d 30公分
300 $a 參考書目：面26-27
300 $a 102年10月31日公開
314 $a 指導教授：郭錕霖博士
328 $a 碩士論文--國立高雄大學統計學研究所
330 $a 我們知道多維常態分佈的條件分佈皆是常態分佈，然而在給定條件分佈是常態的情形下，其相對應的聯合分佈不一定存在（此情形稱為不相容），即使存在（此情形稱為相容）也不一定是常態分佈。在本研究中，我們將探討條件常態分佈模型是相容的充要條件，同時，在相容的情形下，我們將刻劃所對應的聯合分佈之形式。另一方面，有研究者探討在給定的條件分佈模型是不相容的情形下，試圖計算一個聯合分佈，使得此聯合分佈之條件分佈能盡量地接近所給定的條件分佈；也有研究者提出以不相容的條件分配模型進行缺失資料插補，實證效果也有出人意外的表現。在本次研究中，我們將透過模擬的方式，探討上述兩結果。 We know that all conditionals of a multivariate normal distribution are normal distributions. However, if the given conditional distributions are normal, there may be no joint distribution having these conditionals (we call this case to be incompatible). Even though there exists a corresponding joint distribution (we call this case to be compatible), it may not be a normal distribution. In this studying, we will provide the necessary and sufficient conditions for compatibility of conditional normal distribution models. In addition, under compatibility, we will characteristize the corresponding joint distribution.On the other hand, when the given conditional distributions are incompatible, some researchers try to compute one joint distribution whose conditional distributions are as close as possible to the given conditional distributions. In addition, some researchers use incompatible conditional distributions to impute the missing data, and the empirical result is well unexpectedly. In this paper, we will discuss the above outcomes.
510 1 $a Compatibility of Conditional Normal Distribution Model $z eng
610 0 $a 相容性 $a 條件常態分佈 $a 吉氏取樣 $a 偽吉氏分佈
610 1 $a Compatibility $a Conditional Normal Distribution $a Gibbs sampler $a Pseudo-Gibbs distribution
681 $a 008M/0019 $b 343201 8044 $v 2007年版
700 1 $a 曾 $b 婉芸 $4 撰 $3 614517
712 0 2 $a 國立高雄大學 $b 統計學研究所 $3 166081
801 0 $a tw $b NUK $c 20131004 $g CCR
856 7 $z 電子資源 $2 http $u http://handle.ncl.edu.tw/11296/ndltd/45849507622224395914