返回
顶部

软件试用 Nlogit—罗吉特模式软件包

软件简介


NLOGIT是LIMDEP的增强版,为多项式选择数据的评估、模拟、分析提供程序,例如商标选择、运输模式和消费者在一系列竞争中选择的所有形式的测量和市场数据。NLOGIT已经成为评估和模拟多项离散选择模型的首选软件包。
NLOGIT新功能包括:内置超过200个估计量,你可以用来分析
四个层次嵌套logit模型,混合logit随机参数,潜在类别,多项普罗比模型,面板数据——MNL固定效果。

软件功能


NLOGIT是建模的最佳选择
NLOGIT是多项式离散选择模型的评估与模拟方面的标准套装软件。对于其他建模软件,NLOGIT有最大似然估计的全部信息,多达4个级别的嵌入式logit模型。许多其他的公式也包括在NLOGIT里面,如随机参数(混合logit)模型,多项式概率,还有许多形式的嵌式logit模型和一些新的面板数据公式。NLOGIT是唯一一个离散选择分析大型软件包,包含了综合计量经济学软件LIMDEP的全部功能。

数据分析
NLOGIT通常对消费者的个别、横截面数据和多个方案的决策进行分析。但是,它同样可以对市场份额、频率数据、可选择性事物的排名、估计值,以及反复观察得到的面板数据进行分析。除了LIMDEP的分析程序外,NLOGIT还有其他一些处理数据的程序。

模型估计
比起其他任何软件,NLOGIT支持更大范围的离散选择模型。包括基本的多项式logit模型、多达4个级别的嵌入式logit模型,多项式概率模型和最先进的混合logit模型(随机参数)。在所有情况下,能提供多种形式的可用模型。

NLOGIT包含LIMDEP支持的所有离散选择评估模型,并且还包含一些LIMDEP没有的离散选择模型,如:
● 多项式logit — 各种规格的
● MNL随机效应
● 嵌入式logit
● 广义的嵌入式logit
● 多项式概率
● 混合式(随机参数)logit
● 内核logit
● 异方差的极值
● 协方差的异质性
● 潜在类别

模型规格
NLOGIT 估计程序作为LIMDEP模型命令被访问。因为在LIMDEP中,离散选择模型往往比其他单方程模式更加复杂,命令设置包含NLOGIT的许多具体规格。

假设检验的推理工具
NLOGIT可以访问LIMDEP后评估和分析工具的所有功能,包括wald、比率和Lagrange乘数检验以及所有的矩阵代数和科学计算器工具。NLOGIT为离散选择分析提供具体的工具,包括检验多项logit模型IIA假定的内置程序。

模拟
NLOGIT的任何模型估计都可以用模拟功能的“What if”来分析。基准模型为样本数据产生了一组拟合概率,聚合选择集中可选择性的样本股的预测。然后使用模拟器和评估数据集以及其他兼容数据集,来重新计算您指定情况下的股票,比如特别选择价格变化或家庭收入的变化。

报告结果
该模拟器现在可以用来计算和报告弹性。(因为该程序已经设计成用来计算由于属性的离散变化而引起的概率变化,弧弹性就是一个自然延伸的结果)

数据设置和类型
在NLOGIT中,设置多项式离散选择模型的数据主要有3大改进。在NLOGIT模型中,主要数据常常被错误编码或安排不当地分析。而现在您可以要求NLOGIT检查和观察数据,有20种不同的问题会阻碍估计。有些是自动的,其他只要您要求,系统会帮您进行。对于实验性的工作和模型开发,当您提供实用程序时,NLOGIT模型将在类型1极值分布的基础上模拟选择性数据。最后,在某些情况下,调查显示,个人在做出选择时会忽略某些属性。NLOGIT可以自动调节这些数据在模型中的评估。(简单地设置属性为零是不正确的方法 – 比如考虑价格归零。这个模型本身就是可以被修改的。)

NLOGIT的新特征


NLOGIT除了包含LIMDEP 的所有功能外,我们还增加了2个新的模型,在面板数据下的一个嵌入式logit模型和一个等同的随机效应模型。对于分析个别选择来说,混合logit模型(随机参数logit模型)是目前运用最普遍最灵活的模型。新增的模型如下:

广义嵌入式logit模型
嵌入式logit模型是多项logit模型中最流行扩展模型的一种。模型的缺点之一就是严格要求树形结构能准确地把每个选择分配到树结构中的分支中。而广义嵌入式logit模型允许一次在几个分支中出现多个选择和概率。

误差分量Logit 模型
多项式logit模型作为离散选择模型的基本平台已经几十年了。由于不可预测的因素,无法捕捉单个选择的具体变化。误差分量Logit模型弥补了这一缺陷。在重复选择(面板数据)环境下,这一模型将在随机效益模型中的起重要作用。
模型扩展
异质性差异
可以说实用功能的异质性差异跟异质性水平一样重要。我们在容纳异方差到混合logit模型、协方差异质性(嵌入式)模型以及异质性极值模型中增加了具体的规格。

多项式logit模型(GME)
广义上讲最大熵估计提供了一种校准参数的方法,也就是在很多例子中,比起最大似然法,跟数据模型的“fit”更紧密。我们在多项式logit模型和条件logit模型中增加了一个GME评估器,也就是基础MNL模型的所有形式。

混合Logit 模型
如前所述,混合logit模型代表着多项选择模型的前沿。我们增加了许多新的功能以保证这种模型的NLOGIT的实施。其中包括:
增加了许多规则来建立实际的、合理的参数分布。比如,Weibull和三角分布为标记制约系数提供了可行的替代方式。我们还在随机参数的定义中建立了可选的规格,允许出现在不同分布的标准偏差和均值的特征变化。
提供随机系数的异质性的差异(异方差)
误差分量logit 模型可能在混合logit模型的顶部进行分层。
除了对个别具体期望随机参数估计之外,您现在还可以计算支付意愿的具体措施,而这些措施是作为比例系数来计算的。
现在混合logit模型可能适合排列数据。

软件简介(英文)


NLOGIT: Superior Statistical Analysis Software
Complete Statistical Analysis Tools
NLOGIT includes all the features and capabilities of LIMDEP 11 plus NLOGIT’s estimation and analysis tools for multinomial choice modeling.
NLOGIT software is the only large package for choice modeling that contains the full set of features of an integrated statistics program.

The Power of NLOGIT
NLOGIT provides programs for estimation, simulation and analysis of multinomial choice data, such as brand choice, transportation mode, and all manner of survey and market data in which consumers choose among a set of competing alternatives. Since its introduction nearly 20 years ago, NLOGIT has become the premier statistical package for estimation and simulation of multinomial logit models including willingness to pay and best/worst modeling. NLOGIT is the only program available that supports mixing stated and revealed choice data sets.



Superior Analysis Tools for Multinomial Choice Modeling
Our NLOGIT statistical software provides the widest and deepest array of tools available anywhere for analysis of multinomial logit models, including nested logit, generalized mixed multinomial logit, heteroscedastic extreme value, multinomial probit, mixed logit and more. A unique simulation package that allows you to analyze alternative scenarios in the context of any estimated discrete choice model with any data set, whether used in estimation or as hold out data for examining model cross validity.

Function list


Data Analysis
NLOGIT will typically be used to analyze individual, cross section data on consumer choices and decisions from multiple alternatives. But, the program is equally equipped for market shares or frequency data, data on rankings of alternatives, and, for several of the estimators, panel data from repeated observation of choice situations. There are several data handling procedures for NLOGIT in addition to all those available in LIMDEP.

Model Estimation
NLOGIT supports a greater range of models for discrete choice than any other package. These include the basic multinomial logit model, nested logit models up to four levels, the multinomial probit model and the state of the art estimator for the mixed (random parameters) logit model. In all cases, there are a variety of different forms of the model available.
NLOGIT contains all of the discrete choice estimators supported by LIMDEP, plus the extensions of the discrete choice models which do not appear in LIMDEP. These include:

● Multinomial logit - many specifications
● Random effects MNL
● Nested logit
● Generalized nested logit
● Multinomial probit
● Mixed (random parameters) logit
● Kernel logit
● Heteroscedastic extreme value
● Covariance heterogeneity
● Latent class

Model Specification
NLOGIT's estimation programs are accessed as LIMDEP model commands. Since discrete choice models are often more complicated to specify than other single equation models in LIMDEP, the command setup includes many specifications that are specific to NLOGIT.

Inference Tools for Hypothesis Testing
The full set of post estimation and analysis tools in LIMDEP is accessed by NLOGIT. This includes the Wald, likelihood ratio and Lagrange multiplier tests as well as all the matrix algebra and scientific calculator tools. NLOGIT also provides tools specific for discrete choice analysis, including a built-in procedure for testing the IIA assumption of the multinomial logit model.

Simulation
Any model estimated by NLOGIT can be subjected to ‘what if' analyses using the model simulation package. The base case model produces a set of fitted probabilities for the sample data which aggregate to a prediction of the sample shares for the alternatives in the choice set. The simulator is then used, with the estimation data set or any other compatible data set, to recompute these shares under scenarios that you specify, such as a change in the price of a particular alternative or a change in household incomes.

Features of NLOGIT include:
With over 200 built-in estimators, you can analyze:
● Four level nested logit models
● Random parameters mixed logit
● Latent class
● Multinomial probit
● Panel data - fixed effects MNL
● Stated choice experiments
● Willingness to pay
● Heteroscedastic extreme value
● Best/worst modeling
● Random regret
● Attribute nonattendance
● Estimation and simulation and much more

In addition, there are many new features in Version .We have added several enhancements to give you greater flexibility in analyzing different types of data. Many of the features of NLOGIT, existing and new, are designed to let you go beyond just computing coefficients, to analyzing and using your model. We have added many new models including the random regret logit model and best/worst outcome. NLOGIT 6 continues to pioneer new developments for estimation in WTP (willingness to pay) space. Altogether, we have added dozens of features in NLOGIT 6, some clearly visible ones such as the new models and some ‘behind the scenes’ that will smooth the operation and help to stabilize the estimation programs. The following will summarize the important new developments.





New Multinomial Choice Models
NLOGIT includes many new commands and extension of the random parameters model and latent class models:
• Fixed effects in multinomial logit models
• Random effects multinomial logit models
• Random regret logit model
• Best/worst outcome data
• Berry, Levinsohn and Pakes random parameters logit model
• Latent classes with random parameters
• Generalized mixed logit







Model Extensions
• Willingness to pay
• Attribute nonattendance (explicit and implicit)
• Individual specific expected parameters
• Model simulation
• Estimated elasticities and partial effects
• Robust covariance matrix
• Random data generators
• Posterior estimates from latent class models
• Coefficients in random parameters models
• Simplified WALD command