Discriminant Analysis for Sprague-Dawley Rats with Sepsis
碩士 === 逢甲大學 === 統計與精算所 === 96 === The linear discriminant analysis can only deal with such continuous variables problems such as multivariate normality,etc. The assumption of multivariate normality distribution is untenable when the discriminant variable (independent variable) mixed continuous and d...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2008
|
Online Access: | http://ndltd.ncl.edu.tw/handle/75911634485199929479 |
id |
ndltd-TW-096FCU05336003 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-096FCU053360032016-05-18T04:13:56Z http://ndltd.ncl.edu.tw/handle/75911634485199929479 Discriminant Analysis for Sprague-Dawley Rats with Sepsis 敗血症大鼠之判別分析 Chung-Lung Tsai 蔡忠龍 碩士 逢甲大學 統計與精算所 96 The linear discriminant analysis can only deal with such continuous variables problems such as multivariate normality,etc. The assumption of multivariate normality distribution is untenable when the discriminant variable (independent variable) mixed continuous and discrete. In this situation we only can use logistic regression to discrimanant because the logistic regression doesn’t assume any assumptions for the probability distribution of the independent variable. It’s also used in discriminant analysis because logistic regression can used to forecast the successful probability of binary response variables. In this essay, The data of septicaemia experiment for mice is abnormal and repeated , so we can use logistic regression model . Because every mouse’s data of the blood is measured by every two hours, data is not independ. We can use traditional logistic regression to analyse data; otherwise, we also use the logistic normal regression model. We also want to compare these two models. In conclusion, to divide rats into two groups, group of death and group of survival, and predict the death or survive of rats. The discriminating faulty rates of logistic normal regression model is greater than both faulty rates are higher. And those two methods show that the faulty rates are lower when we predict whether the rates are going to die. And we discovered that faulty rates of predicting by using three time points are lower than faulty rates of predicting by using two time points in this experiment. However, in the analysis of three time points, the result of the death group such as two time points is discriminated that thirteen rats are dead and three rats are survive. The discriminating faulty rate is lower than two time points because the observation of survival group is large. Therefore, the total discrimination faulty rate is lower than the two time points. And in fact, using three time points does not have help to predict the death in discriminating. Mei-Jih Gee 紀美智 2008 學位論文 ; thesis 50 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 逢甲大學 === 統計與精算所 === 96 === The linear discriminant analysis can only deal with such continuous variables problems such as multivariate normality,etc. The assumption of multivariate normality distribution is untenable when the discriminant variable (independent variable) mixed continuous and discrete. In this situation we only can use logistic regression to discrimanant because the logistic regression doesn’t assume any assumptions for the probability distribution of the independent variable. It’s also used in discriminant analysis because logistic regression can used to forecast the successful probability of binary response variables.
In this essay, The data of septicaemia experiment for mice is abnormal and repeated , so we can use logistic regression model . Because every mouse’s data of the blood is measured by every two hours, data is not independ. We can use traditional logistic regression to analyse data; otherwise, we also use the logistic normal regression model. We also want to compare these two models.
In conclusion, to divide rats into two groups, group of death and group of survival, and predict the death or survive of rats. The discriminating faulty rates of logistic normal regression model is greater than both faulty rates are higher. And those two methods show that the faulty rates are lower when we predict whether the rates are going to die. And we discovered that faulty rates of predicting by using three time points are lower than faulty rates of predicting by using two time points in this experiment. However, in the analysis of three time points, the result of the death group such as two time points is discriminated that thirteen rats are dead and three rats are survive. The discriminating faulty rate is lower than two time points because the observation of survival group is large. Therefore, the total discrimination faulty rate is lower than the two time points. And in fact, using three time points does not have help to predict the death in discriminating.
|
author2 |
Mei-Jih Gee |
author_facet |
Mei-Jih Gee Chung-Lung Tsai 蔡忠龍 |
author |
Chung-Lung Tsai 蔡忠龍 |
spellingShingle |
Chung-Lung Tsai 蔡忠龍 Discriminant Analysis for Sprague-Dawley Rats with Sepsis |
author_sort |
Chung-Lung Tsai |
title |
Discriminant Analysis for Sprague-Dawley Rats with Sepsis |
title_short |
Discriminant Analysis for Sprague-Dawley Rats with Sepsis |
title_full |
Discriminant Analysis for Sprague-Dawley Rats with Sepsis |
title_fullStr |
Discriminant Analysis for Sprague-Dawley Rats with Sepsis |
title_full_unstemmed |
Discriminant Analysis for Sprague-Dawley Rats with Sepsis |
title_sort |
discriminant analysis for sprague-dawley rats with sepsis |
publishDate |
2008 |
url |
http://ndltd.ncl.edu.tw/handle/75911634485199929479 |
work_keys_str_mv |
AT chunglungtsai discriminantanalysisforspraguedawleyratswithsepsis AT càizhōnglóng discriminantanalysisforspraguedawleyratswithsepsis AT chunglungtsai bàixuèzhèngdàshǔzhīpànbiéfēnxī AT càizhōnglóng bàixuèzhèngdàshǔzhīpànbiéfēnxī |
_version_ |
1718271806882709504 |