During the last thirty eight years, the LISREL model, methods and software have become synonymous with structural equation modeling (SEM). SEM allows researchers in the social sciences, management sciences, behavioral sciences, biological sciences, educational sciences and other fields to empirically assess their theories. These theories are usually formulated as theoretical models for observed and latent (unobservable) variables. If data are collected for the observed variables of the theoretical model, the LISREL program can be used to fit the model to the data.
Today, however, LISREL for Windows is no longer limited to SEM. The latest LISREL for Windows includes the following statistical applications.
LISREL 用于處理結構方程模型(structural equation modeling)。
PRELIS 用于基本的數據處理和簡單的統計分析。
MULTILEV 用于線性和非線性階層模型分析(hierarchical linear and non-linear modeling)。
SURVEYGLIM 用于廣義線性模型(generalized linear modeling)的分析和處理。.
MAPGLIM 用于廣義分層線性模型(generalized linear modeling for multilevel data)的處理。
32位應用程序LISREL被用于:
這些方法可用于以下數據類型:
完整和不完整的復雜調查數據分類和連續變量
完整和不完整的簡單隨機樣本數據分類和連續變量
PRELIS是一個32位的應用程序,可用于:
數據操作
數據轉換
數據生成
矩陣的計算
樣本矩陣的漸近協方差計算
匹配填充
多重填充
多重線性回歸
邏輯回歸
單變量和多變量回歸審查
多線與極小參差法的探索性因素分析
MULTILEV適合于多級線性與非線性模型的簡單隨機和復雜多級數據調查設計,支持模型中包含連續和分類響應變量。
SURVEYGLIM適用于廣義線性模型的簡單隨機和復雜多級數據調查設計,模型可用于以下類型的抽樣分布:
多項分布
伯努利分布
二項分布
負二項分布
泊松分布
正常分布
γ分布
逆高斯分布
MAPGLIM可通過實施最大后驗法(Maximum A Posteriori,MAP)將廣義線性模型與多層次數據進行擬合。
LISREL
The 32-bit application LISREL is intended for:
-
Standard structural equation modeling
-
Multilevel structural equation modeling
These methods are available for the following data types:
-
Complete and incomplete complex survey data on categorical and continuous variables
-
Complete and incomplete simple random sample data on categorical and continuous variables
PRELIS
PRELIS is a 32-bit application which can be used for:
-
Data manipulation
-
Data transformation
-
Data generatiion
-
Computing moment matrices
-
Computing asymptotic covariance matrices of sample moments
-
Imputation by matching
-
Multiple imputation
-
Multiple linear regression
-
Logistic regression
-
Univariate and multivariate censored regression
-
ML and MINRES exploratory factor analysis
MULTILEV
MULTILEV fits multilevel linear and nonlinear models to multilevel data from simple random and complex survey designs. It allows for models with continuous and categorical response variables.
SURVEYGLIM
SURVEYGLIM fits Generalized LInear Models (GLIMs) to data from simple random and complex survey designs.
Models for the following sampling distributions are available.
-
Multinomial
-
Bernoulli
-
Binomial
-
Negative Binomial
-
Poisson
-
Normal
-
Gamma
-
Inverse Gaussian
MAPGLIM
MAPGLIM implements the Maximum A Priori (MAP) method to fit generalized linear models to multilevel data.