Version info: Code for this page was tested in SPSS 20.

Logistic regression, also called a logit model, is used to model dichotomous outcome variables. In the logit model the log odds of the outcome is modeled as a linear combination of the predictor variables.

Please note: The purpose of this page is to show how to use various data analysis commands. It does not cover all aspects of the research process which researchers are expected to do. In particular, it does not cover data cleaning and checking, verification of assumptions, model diagnostics and potential follow-up analyses.

Examples

Example 1:  Suppose that we are interested in the factors

that influence whether a political candidate wins an election.  The

outcome (response) variable is binary (0/1);  win or lose.

The predictor variables of interest are the amount of money spent on the campaign, the

amount of time spent campaigning negatively and whether or not the candidate is an

incumbent.

Example 2:  A researcher is interested in how variables, such as GRE (Graduate Record Exam scores),

GPA (grade

point average) and prestige of the undergraduate institution, effect admission into graduate

school. The response variable, admit/don’t admit, is a binary variable.

Description of the data

For our data analysis below, we are going to expand on Example 2 about getting into graduate school.  We have generated hypothetical data, which can be obtained from our website by clicking on binary.sav. You can store this anywhere you like, but the syntax below assumes it has been stored in the directory c:data. This dataset has a binary response (outcome, dependent) variable called admit, which is equal to 1 if the individual was admitted to graduate school, and 0 otherwise. There are three predictor variables: gre, gpa, and rank. We will treat the variables gre and gpa as continuous. The variable rank takes on the values 1 through 4. Institutions with a rank of 1 have the highest prestige, while those with a rank of 4 have the lowest. We start out by opening the dataset and looking at some descriptive statistics.

get file = "c:databinary.sav".

descriptives /variables=gre gpa.


frequencies /variables = rank.


crosstabs /tables = admit by rank.

Analysis methods you might consider

Below is a list of some analysis methods you may have encountered. Some of the methods listed are quite reasonable while others have either fallen out of favor or have limitations.

  • Logistic regression, the focus of this page.
  • Probit regression.  Probit analysis will produce results similar

    logistic regression. The choice of probit versus logit depends largely on

    individual preferences.

  • OLS regression.  When used with a binary response variable, this model is known

    as a linear probability model and can be used as a way to

    describe conditional probabilities. However, the errors (i.e., residuals) from the linear probability model violate the homoskedasticity and

    normality of errors assumptions of OLS

    regression, resulting in invalid standard errors and hypothesis tests. For

    a more thorough discussion of these and other problems with the linear

    probability model, see Long (1997, p. 38-40).

  • Two-group discriminant function analysis. A multivariate method for dichotomous outcome variables.
  • Hotelling’s T2.  The 0/1 outcome is turned into the

    grouping variable, and the former predictors are turned into outcome

    variables. This will produce an overall test of significance but will not

    give individual coefficients for each variable, and it is unclear the extent

    to which each "predictor" is adjusted for the impact of the other

    "predictors."

Logistic regression

Below we use the logistic regression command to run a model predicting the outcome variable admit, using gre, gpa, and rank. The categorical option specifies that rank is a categorical rather than continuous variable. The output is shown in sections, each of which is discussed below.

logistic regression admit with gre gpa rank
/categorical = rank.

The first table above shows a breakdown of the number of cases used and not used in the analysis. The second table above gives the coding for the outcome variable, admit.

The table above shows how the values of the categorical variable rank were handled, there are terms (essentially dummy variables) in the model for rank=1,rank=2, and rank=3; rank=4 is the omitted category.

  • The first model in the output is a null model, that is, a model with no predictors.
  • The constant in the table labeled Variables in the Equation gives the unconditional log odds of admission (i.e., admit=1).
  • The table labeled Variables not in the Equation gives the results of a score test, also known as a Lagrange multiplier test. The column labeled Score gives the estimated change in model fit if the term is added to the model, the other two columns give the degrees of freedom, and p-value (labeled Sig.) for the estimated change. Based on the table above, all three of the predictors, gre, gpa, and rank, are expected to improve the fit of the model.

  • The first table above gives the overall test for the model that includes the predictors. The chi-square value of 41.46 with a p-value of less than 0.0005 tells us that our model as a whole fits significantly better than an empty model (i.e., a model with no predictors).
  • The -2*log likelihood (458.517) in the Model Summary table can be used in comparisons of nested models, but we won’t show an example of that here. This table also gives two measures of pseudo R-square.

  • In the table labeled Variables in the Equation we see the coefficients, their standard errors, the Wald test statistic with associated  degrees of freedom and p-values, and the exponentiated coefficient (also known as an odds ratio).  Both gre and gpa are statistically significant. The overall (i.e., multiple degree of freedom) test for rank is given first, followed by the terms for rank=1, rank=2, and rank=3. The overall effect of rank is statistically significant, as are the terms for rank=1 and rank=2. The logistic regression coefficients give the change in the log odds of the outcome for a one unit increase in the predictor variable.

    • For every one unit change in gre, the log odds of admission (versus non-admission) increases by 0.002.
    • For a one unit increase in gpa, the log odds of being admitted to graduate school increases by 0.804.
    • The indicator variables for rank have a slightly different interpretation. For example, having attended an undergraduate institution with rank of 1, versus an institution with a rank of 4, increases the log odds of admission by 1.551.

Things to consider

  • Empty cells or small cells:  You should check for empty or small

    cells by doing a crosstab between categorical predictors and the outcome variable.  If a cell has very few cases (a small cell), the model may become unstable or it might not run at all.

  • Separation or quasi-separation (also called perfect prediction), a condition in which the outcome does not vary at some levels of the independent variables. See our page FAQ: What is complete or quasi-complete separation in logistic/probit regression and how do we deal with them? for information on models with perfect prediction.
  • Sample size:  Both logit and probit models require more cases than OLS regression because they use maximum likelihood estimation techniques.  It is also important to keep in mind that when the outcome is rare, even if the overall dataset is large, it can be difficult to estimate a logit model.
  • Pseudo-R-squared:  Many different measures of pseudo-R-squared exist. They all attempt to provide information similar to that provided by R-squared in OLS regression; however, none of them can be interpreted exactly as R-squared in OLS regression is interpreted. For a discussion of various pseudo-R-squareds see Long and Freese (2006) or our FAQ page What are pseudo R-squareds?
  • Diagnostics:  The diagnostics for logistic regression are different from those for OLS regression. For a discussion of model diagnostics for logistic regression, see Hosmer and Lemeshow (2000, Chapter 5). Note that diagnostics done for logistic regression are similar to those done for probit regression.

See also

References

  • Hosmer, D. & Lemeshow, S. (2000). Applied Logistic Regression (Second Edition).

    New York: John Wiley & Sons, Inc.

  • Long, J. Scott (1997). Regression Models for Categorical and Limited Dependent Variables.

    Thousand Oaks, CA: Sage Publications.

LOGIT REGRESSION的更多相关文章

  1. Logistic回归模型和Python实现

    回归分析是研究变量之间定量关系的一种统计学方法,具有广泛的应用. Logistic回归模型 线性回归 先从线性回归模型开始,线性回归是最基本的回归模型,它使用线性函数描述两个变量之间的关系,将连续或离 ...

  2. 机器学习算法基础(Python和R语言实现)

    https://www.analyticsvidhya.com/blog/2015/08/common-machine-learning-algorithms/?spm=5176.100239.blo ...

  3. Day3 《机器学习》第三章学习笔记

    这一章也是本书基本理论的一章,我对这章后面有些公式看的比较模糊,这一会章涉及线性代数和概率论基础知识,讲了几种经典的线性模型,回归,分类(二分类和多分类)任务. 3.1 基本形式 给定由d个属性描述的 ...

  4. 从线性模型(linear model)衍生出的机器学习分类器(classifier)

    1. 线性模型简介 0x1:线性模型的现实意义 在一个理想的连续世界中,任何非线性的东西都可以被线性的东西来拟合(参考Taylor Expansion公式),所以理论上线性模型可以模拟物理世界中的绝大 ...

  5. 壁虎书4 Training Models

    Linear Regression The Normal Equation Computational Complexity 线性回归模型与MSE. the normal equation: a cl ...

  6. python: 模型的统计信息

    /*! * * Twitter Bootstrap * */ /*! * Bootstrap v3.3.7 (http://getbootstrap.com) * Copyright 2011-201 ...

  7. [译]用R语言做挖掘数据《四》

    回归 一.实验说明 1. 环境登录 无需密码自动登录,系统用户名shiyanlou,密码shiyanlou 2. 环境介绍 本实验环境采用带桌面的Ubuntu Linux环境,实验中会用到程序: 1. ...

  8. Machine Learning and Data Mining(机器学习与数据挖掘)

    Problems[show] Classification Clustering Regression Anomaly detection Association rules Reinforcemen ...

  9. (数据科学学习手札24)逻辑回归分类器原理详解&Python与R实现

    一.简介 逻辑回归(Logistic Regression),与它的名字恰恰相反,它是一个分类器而非回归方法,在一些文献里它也被称为logit回归.最大熵分类器(MaxEnt).对数线性分类器等:我们 ...

随机推荐

  1. 数据统计,包括mysql和MongoDB

    select ct.dt, COUNT(DISTINCT c.id) from tms_service_customer c, tms_dispatch_details d, (select DIST ...

  2. Codeforces Round #587 (Div. 3) D. Swords

    链接: https://codeforces.com/contest/1216/problem/D 题意: There were n types of swords in the theater ba ...

  3. Codeforces Round #588 (Div. 2) D. Marcin and Training Camp(思维)

    链接: https://codeforces.com/contest/1230/problem/D 题意: Marcin is a coach in his university. There are ...

  4. ISCSI共享存储

    ISCSI网络磁盘    默认端口:3260 服务端: 一. 二.安装软件:targetcli 用命令targetcli进行配置------------------------进入iscsi磁盘配置模 ...

  5. jQuery系列(十二):事件委托

    1.什么是事件委托 通俗的讲,事件就是onclick,onmouseover,onmouseout,等就是事件,委托呢,就是让别人来做,这个事件本来是加在某些元素上的,然而你却加到别人身上来做,完成这 ...

  6. C/C++程序基础-标准头文件的结构

    1:标准头文件有如下结构,请解释. #ifndef _INCvxWorksh #define _INCvxWorksh#ifdef _cplusplus extern "C"{ # ...

  7. 第11组 团队Git现场编程实战

    第11组 团队Git现场编程实战 组员职责分工: 前端部分: 陈郑铧:构架的搭建,前端模块开发 陈益:前端模块开发 李镇平:前端模块开发 后端部分: 沈国煜:后端模块开发 王泽鸿:后端模块开发 林铮威 ...

  8. 网络流,设备、插头和转接器建图(简单map的应用)

    题意: 给你n个插座,m个设备,每台设备都有对应的插座,有k个转接器. 要求:求满足不能插上插座的用电器最少个数 solution: HINT:每种适配器都有无限个,所以建图的时候要改为INF. 答案 ...

  9. 之前有面试到两个日期的大小比较方式,现在整理一下几种方法。   例子:   String beginTime=new String("2017-06-09 10:22:22");     String endTime=new String("2017-05-08 11:22:22");  1  直接用Date自带方法before()和after()比较 SimpleDateFormat d

    各种数据类型(日期/时间.integer.floating point和numeric)转换成格式化的字符串以及反过来从格式化的字符串转换成指定的数据类型.下面列出了这些函数,它们都遵循一个公共的调用 ...

  10. Java-CAS 与原子类

    CAS(Compare and Swap),即比较并替换,实现并发算法时常用到的一种技术. CAS 的思想很简单:三个参数,一个当前内存值 V.旧的预期值 A.即将更新的值 B,当且仅当预期值 A 和 ...