Estimation of non-Gaussian SVAR models: a pseudo-log-likelihood function approach
Estimation of non-Gaussian SVAR models: a pseudo-log-likelihood function approach
复制标题
DOI：
10.1080/00949655.2022.2155160
复制
发表时间：
2022-12
影响因子：
1.2
通讯作者：
K. Maekawa;Tadashi Nakanishi
中科院分区：
文献类型：
--
作者：
K. Maekawa;Tadashi Nakanishi
本文研究了扰动具有非高斯分布的结构向量自回归模型的参数估计问题。我们称之为非高斯向量自回归（NG-SVAR）模型。由于该模型的估计问题与机器学习和信号处理中的独立分量分析（伊卡）密切相关，因此我们将伊卡理论应用于我们的估计问题。然而，由于我们不知道实际上真正的非高斯分布，我们无法构造精确的对数似然函数。在本文中，我们提出了一个伪最大对数似然估计。从半参数统计的角度来看，我们的估计是统计有效的。此外，我们通过Monte Carlo实验和小样本下的实证例子证明了我们的估计具有令人满意的性能。
We consider estimation problem in structural vector autoregressive model which disturbance has non-Gaussian distribution. We call this model as non-Gaussian vector autoregressive (NG-SVAR) model. Since the estimation problem of this model is closely related to the independent component analysis (ICA) developed in machine learning and signal processing we apply the theory of ICA to our estimation problem. However, since we do not know the true non-Gaussian distribution in practice, we cannot construct the exact loglikelihood function. In this paper we propose a pseudo maximum loglikelihood estimator instead. It is shown that our estimator is statistical efficient from view point of semiparametric statistics. Furthermore, we show that our estimator has satisfactory performance by Monte Carlo experiment and empirical example in small sample.