arch {vars}

R Documentation

ARCH-LM testDescriptionThis function computes univariate and multivariate ARCH-LM tests for aVAR(p).
Usagearch(x, lags.single = 16, lags.multi = 5, multivariate.only = TRUE)
Arguments

x	Object of class ‘varest’; generated byVAR(), or an object of class ‘vec2var’; generated byvec2var().
lags.single	An integer specifying the lags to be used for theunivariate ARCH statistics.
lags.multi	An integer specifying the lags to be used for themultivariate ARCH statistic.
multivariate.only	If TRUE (the default), onlythe multivariate test statistic is computed.

DetailsThe multivariate ARCH-LM test is based on the following regression(the univariate test can be considered as special case of theexhibtion below and is skipped):

vech(hat{u}_t hat{u}_t') = β_0 + B_1vech(hat{u}_{t-1} hat{u}_{t-1}') + ... + B_qvech(hat{u}_{t-q} hat{u}_{t-q}' + v_t)

whereby v_t assigns a spherical error process andvech is the column-stacking operator for symmetric matricesthat stacks the columns from the main diagonal on downwards. Thedimension of β_0 is frac{1}{2}K(K +1) and forthe coefficient matrices B_i with i=1, ..., q,frac{1}{2}K(K +1) times frac{1}{2}K(K +1). The nullhypothesis is: H_0 := B_1 = B_2 = ... = B_q = 0 and thealternative is: H_1: B_1 neq 0 or B_2 neq 0 or ... B_q  neq0.The test statistic is:

VARCH_{LM}(q) = frac{1}{2}T K (K + 1)R_m^2 quad ,

with

R_m^2 = 1 - frac{2}{K(K+1)}tr(hat{Omega} hat{Omega}_0^{-1})quad ,

and hat{Omega} assigns the covariance matrix of the abovedefined regression model. This test statistic is distributed aschi^2(qK^2(K+1)^2/4).
ValueA list with class attribute ‘varcheck’ holding thefollowing elements:

resid	A matrix with the residuals of the VAR.
arch.uni	A list with objects of class ‘htest’containing the univariate ARCH-LM tests per equation. This elementis only returned if multivariate.only = FALSE is set.
arch.mul	An object with class attribute ‘htest’containing the multivariate ARCH-LM statistic.

Author(s)Bernhard Pfaff
ReferencesDoornik, J. A. and D. F. Hendry (1997), Modelling DynamicSystems Using PcFiml 9.0 for Windows, International ThomsonBusiness Press, London.
Engle, R. F. (1982), Autoregressive conditional heteroscedasticitywith estimates of the variance of United Kingdom inflation,Econometrica, 50: 987-1007.
Hamilton, J. (1994), Time Series Analysis, PrincetonUniversity Press, Princeton.
L

这个。。。楼上的基本就是抄help吧，呵呵

是啊,不过不是基本抄,简直就是完全的复制粘贴.希望能有帮助,钱是不要的.

用专门做金融时间序列的FinTS包，其中ArchTest（）函数就是用来做Arch LM检验的
例子：
dat <- ArchTest(x, lags=12, demean = FALSE)
参数解释：
x                numeric vector  #数值型向量
lags                positive integer number of lags #滞后阶数
demean        logical: If TRUE, remove the mean before computing the test statistic. #是否做中心化处理
可以参考蔡瑞胸的金融时间序列分析一书哦~

那你说的，对面板数据怎么检验啊？解释变量有四五个呢