Maximum likelihood estimation has been widely applied in system identification because of consistency, its asymptotic efficiency and sufficiency. However, gradient-based optimisation of the likelihood function might end up in local convergence. In this article we derive various new non-local-minimum conditions in both open and closed-loop system when the noise distribution is a Gaussian process. Here we consider different model structures, in particular ARARMAX, BJ and OE models.View full textDownload full textKeywordsMLE, optimisation, global/local convergence, consistency, asymptotic efficiency and sufficiency, non-local-minimum conditionsRelated var addthis_config = { ui_cobrand: "Taylor & Francis Online", services_compact: "citeulike,netvibes,twitter,technorati,delicious,linkedin,facebook,stumbleupon,digg,google,more", pubid: "ra-4dff56cd6bb1830b" }; Add to shortlist Link Permalink http://dx.doi.org/10.1080/00207179.2012.658085
展开▼
机译：由于一致性，渐近效率和充分性，最大似然估计已广泛应用于系统识别。但是，似然函数的基于梯度的优化可能会导致局部收敛。在本文中，当噪声分布是高斯过程时，我们可以得出开环和闭环系统中的各种新的非局部最小条件。这里我们考虑不同的模型结构，尤其是ARARMAX，BJ和OE模型。查看全文下载全文关键字MLE，优化，全局/局部收敛，一致性，渐近效率和充分性，非局部最小条件相关的var addthis_config = {ui_cobrand：“ Taylor ＆Francis Online”，services_compact：“ citeulike，netvibes，twitter，technorati，delicious，linkedin，facebook，stumbleupon，digg，google，更多”，发布号：“ ra-4dff56cd6bb1830b”};添加到候选列表链接永久链接http://dx.doi.org/10.1080/00207179.2012.658085
展开▼
