lnam(sna)
lnam()所属R语言包：sna
Fit a Linear Network Autocorrelation Model
装一个线性网络自相关模型
译者：生物统计家园网 机器人LoveR
描述----------Description----------
lnam is used to fit linear network autocorrelation models. These include standard OLS as a special case, although lm is to be preferred for such analyses.
lnam线性网络自相关模型来拟合。这些措施包括标准的OLS作为一个特殊的情况下，虽然lm这样的分析是首选。
用法----------Usage----------
lnam(y, x = NULL, W1 = NULL, W2 = NULL, theta.seed = NULL,
null.model = c("meanstd", "mean", "std", "none"), method = "BFGS",
control = list(), tol=1e-10)
参数----------Arguments----------
参数：y
a vector of responses.
一个向量的响应。
参数：x
a vector or matrix of covariates; if the latter, each column should contain a single covariate.
一个向量或矩阵的协变量，如果是后者，每列包含一个单一的协变量。
参数：W1
one or more (possibly valued) graphs on the elements of y.
一个或多个（可能值）图表元素y。
参数：W2
one or more (possibly valued) graphs on the elements of y.
一个或多个（可能值）图表元素y。
参数：theta.seed
an optional seed value for the parameter vector estimation process.
一个可选的种子值的参数向量的估计处理。
参数：null.model
the null model to be fit; must be one of "meanstd", "mean", "std", or "none".
必须是一个空模型是合适的;的"meanstd"，"mean"，"std"或"none"。
参数：method
method to be used with optim.
的方法，可以使用与optim。
参数：control
optional control parameters for optim.
可选的控制参数为optim。
参数：tol
convergence tolerance for the MLE (expressed as change in deviance).
收敛误差极大似然估计的（表示变化偏差）。
Details
详细信息----------Details----------
lnam fits the linear network autocorrelation model given by
lnam符合线性网络自相关模型
y = W1 %*% y + X %*% beta + e, e = W2 %*% e + nu</i>
Y = W1％*％Y + X％*％β+ E，E = W2％*％E + NU </ I>
where y is a vector of responses, X is a covariate matrix, nu ~ Norm(0,sigma^2),
y是一个向量的反应，X是协变量矩阵，nu ~ Norm(0,sigma^2)，
W1 = sum( rho1_i W1_i, i=1..p), W2 = sum( rho2_i W2_i, i=1..q),</i>
W1 =的的总和（rho1_i W1_i，i = 1 .. P），W2 = SUM（rho2_i W2_i，i = 1 .. Q）</ I>
and W1_i, W2_i are (possibly valued) adjacency matrices.
和W1_i，W2_i（可能值）邻接矩阵。
Intuitively, rho1 is a vector of “AR”-like parameters (parameterizing the autoregression of each y value on its neighbors in the graphs of W1) while rho2 is a vector of “MA”-like parameters (parameterizing the autocorrelation of each disturbance in y on its neighbors in the graphs of W2). In general, the two models are distinct, and either or both effects may be selected by including the appropriate matrix arguments.
直观地说，rho1是一个向量，“AR”般的参数（参数自回归的每一个y对邻国的图形W1值），而rho2是一个向量的“MA”般的参数（每个干扰参数的自相关yW2）在图中的邻居。在一般情况下，这两个模型是不同的，并包括适当的矩阵参数的选择可以通过任一或两者的效果。
Model parameters are estimated by maximum likelihood, and asymptotic standard errors are provided as well; all of the above (and more) can be obtained by means of the appropriate print and summary methods. A plotting method is also provided, which supplies fit basic diagnostics for the estimated model. For purposes of comparison, fits may be evaluated against one of four null models:
通过最大似然估计模型参数，渐近标准误差以及上述所有（及以上）可以通过适当的print和summary方法。还提供了绘图方法的估计模型，它提供合适的基本诊断。为了比较的目的，配合可以评价对四个零模型之一：
<ol> meanstd: mean and standard deviation estimated (default).
<OL>meanstd：均值和标准差估计（默认）。
mean: mean estimated; standard deviation assumed equal to 1.
“mean：假设估计，标准差等于1。
std: standard deviation estimated; mean assumed equal to 0.
std：标准偏差估计;，意味着假设等于0。
none: no parameters estimated; data assumed to be drawn from a standard normal density. </ol>
none：没有参数估计，假设得出的数据，从一个标准的正常密度。 </ OL>
The default setting should be appropriate for the vast majority of cases, although the others may have use when fitting “pure” autoregressive models (e.g., without covariates). Although a major use of the lnam is in controlling for network autocorrelation within a regression context, the model is subtle and has a variety of uses. (See the references below for suggestions.)
默认设置应适用于绝大多数情况下，虽然其他人可能已经装修时使用的“纯”自回归模型（例如，没有协变量）。虽然使用的lnam的主要是在控制网络在回归中的自相关，该模型是微妙的，具有多种用途。 （请参阅下面的参考资料的建议。）
值----------Value----------
An object of class "lnam" containing the following elements:
对象的类"lnam"包含以下元素：
<table summary="R valueblock"> <tr valign="top"><td>y</td> <td> the response vector used.</td></tr> <tr valign="top"><td>x</td> <td> if supplied, the coefficient matrix.</td></tr> <tr valign="top"><td>W1</td> <td> if supplied, the W1 array.</td></tr> <tr valign="top"><td>W2</td> <td> if supplied, the W2 array.</td></tr> <tr valign="top"><td>model</td> <td> a code indicating the model terms fit.</td></tr> <tr valign="top"><td>infomat</td> <td> the estimated Fisher information matrix for the fitted model.</td></tr> <tr valign="top"><td>acvm</td> <td> the estimated asymptotic covariance matrix for the model parameters.</td></tr> <tr valign="top"><td>null.model</td> <td> a string indicating the null model fit.</td></tr> <tr valign="top"><td>lnlik.null</td> <td> the log-likelihood of y under the null model.</td></tr> <tr valign="top"><td>df.null.resid</td> <td> the residual degrees of freedom under the null model.</td></tr> <tr valign="top"><td>df.null</td> <td> the model degrees of freedom under the null model.</td></tr> <tr valign="top"><td>null.param</td> <td> parameter estimates for the null model.</td></tr> <tr valign="top"><td>lnlik.model</td> <td> the log-likelihood of y under the fitted model.</td></tr> <tr valign="top"><td>df.model</td> <td> the model degrees of freedom.</td></tr> <tr valign="top"><td>df.residual</td> <td> the residual degrees of freedom.</td></tr> <tr valign="top"><td>df.total</td> <td> the total degrees of freedom.</td></tr> <tr valign="top"><td>rho1</td> <td> if applicable, the MLE for rho1.</td></tr> <tr valign="top"><td>rho1.se</td> <td> if applicable, the asymptotic standard error for rho1.</td></tr> <tr valign="top"><td>rho2</td> <td> if applicable, the MLE for rho2.</td></tr> <tr valign="top"><td>rho2.se</td> <td> if applicable, the asymptotic standard error for rho2.</td></tr> <tr valign="top"><td>sigma</td> <td> the MLE for sigma.</td></tr> <tr valign="top"><td>sigma.se</td> <td> the standard error for sigma</td></tr> <tr valign="top"><td>beta</td> <td> if applicable, the MLE for beta.</td></tr> <tr valign="top"><td>beta.se</td> <td> if applicable, the asymptotic standard errors for beta.</td></tr> <tr valign="top"><td>fitted.values</td> <td> the fitted mean values.</td></tr> <tr valign="top"><td>residuals</td> <td> the residuals (response minus fitted); note that these correspond to e-hat in the model equation, not nu-hat.</td></tr> <tr valign="top"><td>disturbances</td> <td> the estimated disturbances, i.e., nu-hat.</td></tr> <tr valign="top"><td>call</td> <td> the matched call.</td></tr> </table>
<table summary="R valueblock"> <tr valign="top"> <TD> y</ TD> <TD>的响应向量使用。</ TD> </ TR> <TR VALIGN =“顶“<TD> x </ TD> <TD>如果提供的系数矩阵。</ TD> </ TR> <tr valign="top"> <TD>W1 / TD> <TD>如果提供的话，的W1阵列。</ TD> </ TR> <tr valign="top"> <TD>W2 </ TD> <TD>如果提供的话，W2阵列</ TD> </ TR> <tr valign="top"> <TD>model </ TD> <td>一个代码，以显示模型项符合。</ TD> </ TR> <TR VALIGN =“”> <TD>infomat </ TD> <TD>估计的拟合模型的Fisher信息矩阵。</ TD> </ TR> <tr valign="top"> <TD> acvm</ TD> <TD>的模型参数估计的渐近协方差矩阵。</ TD> </ TR> <tr valign="top"> <TD> null.model</ TD> <td>一个字符串，表示空模型拟合。</ TD> </ TR> <tr valign="top"> <TD>lnlik.null </ TD> <TD>的log - y的可能性下空模型。</ TD> </ TR> <tr valign="top"> <TD>df.null.resid </ TD> <TD>的剩余自由度下的空模型。</ TD> </ TR> <tr valign="top"> <TD> df.null</ TD> <TD>模型空模式下的自由程度。</ TD> </ TR> <TR VALIGN =“顶” > <TD> null.param </ TD> <TD>空模型的参数估计值。</ TD> </ TR> <tr valign="top"> <TD>lnlik.model</ TD > <TD> y的拟合模型下的对数似然。</ TD> </ TR> <tr valign="top"> <TD>df.model </ TD> <TD>模型的度自由。</ TD> </ TR> <tr valign="top"> <TD>df.residual </ TD> <TD>的剩余自由度。</ TD> </ TR> <TR VALIGN =“”> <TD>df.total </ TD> <TD>的总自由度。</ TD> </ TR> <tr valign="top"> <TD>rho1 </ TD> <TD>如果适用的话，最大似然估计的RHO1。</ TD> </ TR> <tr valign="top"> <TD>rho1.se</ TD> <TD>（如适用）， </ TD>渐近标准误差RHO1。</ TR> <tr valign="top"> <TD> rho2</ TD> <TD>（如果适用）的MLE为RHO2。</ TD> < / TR> <tr valign="top"> <TD> rho2.se </ TD> <TD>（如适用）的渐近标准误差为RHO2。</ TD> </ TR> <TR VALIGN =“顶部“> <TD> sigma </ TD> <TD>的MLE为Sigma。</ TD> </ TR> <tr valign="top"> <TD> sigma.se </ TD>的标准误差为<TD> SIGMA </ TD> </ TR> <tr valign="top"> <TD> beta </ TD> <TD>（如适用）的极大似然估计的β</ TD > </ TR> <tr valign="top"> <TD> beta.se </ TD> <TD>（如适用）的渐近标准误差测试。</ TD> </ TR> <TR VALIGN = “顶”> <TD> fitted.values </ TD> <TD>装的平均值。</ TD> </ TR> <tr valign="top"> <TD>residuals</ TD> <TD>的残差（响应减的话）;注意，这些对应于e-hat的模型方程，而不是nu-hat。</ TD> </ TR> <TR VALIGN =“顶” > <TD> disturbances </ TD> <TD>估计的干扰，即nu-hat。</ TD> </ TR> <tr valign="top"> <TD><X > </ TD> <TD>匹配的呼叫。</ TD> </ TR> </ TABLE>
注意----------Note----------
Actual optimization is performed by calls to optim. Information on algorithms and control parameters can be found via the appropriate man pages.
实际的优化是通过调用optim。算法和控制参数的信息可以通过相应的手册页。
（作者）----------Author(s)----------
Carter T. Butts <a href="mailto:buttsc@uci.edu">buttsc@uci.edu</a>
参考文献----------References----------
Leenders, T.Th.A.J. (2002) “Modeling Social Influence Through Network Autocorrelation: Constructing the Weight Matrix” Social Networks, 24(1), 21-47.
Anselin, L. (1988) Spatial Econometrics: Methods and Models. Norwell, MA: Kluwer.
参见----------See Also----------
lm, optim
lm，optim
实例----------Examples----------
## Not run: [＃不运行：]
#Construct a simple, random example:[构造一个简单的，随便举个例子：]
w1<-rgraph(100) #Draw the AR matrix[绘制AR矩阵]
w2<-rgraph(100) #Draw the MA matrix[绘制MA矩阵]
x<-matrix(rnorm(100*5),100,5) #Draw some covariates[绘制一些协变量]
r1<-0.2 #Set the model parameters[设置的模型参数]
r2<-0.1
sigma<-0.1
beta<-rnorm(5)
#Assemble y from its components:[组装y从它的成分：]
nu<-rnorm(100,0,sigma) #Draw the disturbances[绘制的骚乱]
e<-qr.solve(diag(100)-r2*w2,nu) #Draw the effective errors[绘制有效的错误]
y<-qr.solve(diag(100)-r1*w1,x%*%beta+e) #Compute y[计算y]
#Now, fit the autocorrelation model:[现在，适合自相关模型：]
fit<-lnam(y,x,w1,w2)
summary(fit)
plot(fit)
## End(Not run)[＃（不执行）]
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注：
注1：为了方便大家学习，本文档为生物统计家园网机器人LoveR翻译而成，仅供个人R语言学习参考使用，生物统计家园保留版权。
注2：由于是机器人自动翻译，难免有不准确之处，使用时仔细对照中、英文内容进行反复理解，可以帮助R语言的学习。
注3：如遇到不准确之处，请在本贴的后面进行回帖，我们会逐渐进行修订。
|
发表于 2012-9-30 10:55:52
