## 统计代写|时间序列分析代写Time-Series Analysis代考|STAT435

2022年9月24日

couryes-lab™ 为您的留学生涯保驾护航 在代写时间序列分析Time-Series Analysis方面已经树立了自己的口碑, 保证靠谱, 高质且原创的统计Statistics代写服务。我们的专家在代写时间序列分析Time-Series Analysis代写方面经验极为丰富，各种代写时间序列分析Time-Series Analysis相关的作业也就用不着说。

• Statistical Inference 统计推断
• Statistical Computing 统计计算
• Advanced Probability Theory 高等概率论
• Advanced Mathematical Statistics 高等数理统计学
• (Generalized) Linear Models 广义线性模型
• Statistical Machine Learning 统计机器学习
• Longitudinal Data Analysis 纵向数据分析
• Foundations of Data Science 数据科学基础
couryes™为您提供可以保分的包课服务

## 统计代写|时间序列分析代写Time-Series Analysis代考|TESTING FOR NONLINEARITY

11.36 As the previous sections have demonstrated, there have been a wide variety of nonlinear processes proposed for modeling time series. We have, for example, compared ARCH and bilinear models, and in so doing have discussed LM tests for each. Nevertheless, given the range of alternative nonlinear models, it is not surprising that other tests for nonlinearity have also been proposed. Since the form of the departure from linearity is often difficult to specify a priori, many tests are diagnostic in nature, i.e., a clear alternative to the null hypothesis of linearity is not specified. This, of course, leads to difficulties in discriminating between the possible causes of any “nonlinear misspecification” that might be uncovered by such tests.
11.37 The detection of nonlinearity is further complicated by the fact that it has similar symptoms to other types of time series behavior. For example, Andersson et al. (1999) have shown that long memory may lead to spurious rejection of the linearity hypothesis. As demonstrated by Granger and Teräsvirta (1999) and Diebold and Inoue (2001), the opposite may also be true, since some nonlinear processes exhibit characteristics that might justify modeling via a long memory model.
11.38 A related approach considers testing and modeling nonlinearity within a long memory process (see, for example, Baillie and Kapetanios, 2007). Koop and Potter (2001) have shown that unpredictable structural instability in a time series may also produce erroneous evidence of threshold-type nonlinearity. An alarming finding by Ghysels et al. (1996) is that nonlinear transformations, such as the X11 seasonal adjustment procedure, that are routinely applied prior to time series modeling, may also induce nonlinear behavior. Equally, seasonal adjustments may smooth out 1999). Finally, as discussed by Van Dijk et al. (1999) and De Lima (1997), neglecting outliers and nonnormalities may also lead to spurious evidence of nonlinearity.

Despite these difficulties, testing for nonlinearity is usually an effort well spent, since the burden associated with the specification and estimation of nonlinear models is often substantial and complex.
11.39 Empirical applications and simulation studies (e.g., Lee et al., 1993; Barnett et al., 1997) have shown that no nonlinearity test dominates in all situations and that power varies with sample size and the characteristics of the underlying stochastic process. This means that, in practice, it is advisable to apply a variety of nonlinearity tests to the data to guide the model specification process.
11.40 Based on Volterra expansions, Ramsey (1969), Keenan (1985), and Tsay (1986b) provide regression-type tests of linearity against unspecified alternatives. These appear to have good power against the nonlinear moving average and bilinear alternatives, but possibly low power against ARCH models. In developing these tests, we assume that an $\operatorname{AR}(p)$ process has been fitted to the observed series $x_t$ and that the residuals, $e_t$, and the fitted values, $\hat{x}_t=x_t-e_t$, have been calculated.

## 统计代写|时间序列分析代写Time-Series Analysis代考|TRANSFER FUNCTION-NOISE MODELS

12.1 The models that have been developed so far in this book have all been univariate, so that the current value of a time series depends, linearly or otherwise, only on past values of itself and, perhaps, a deterministic function of time. While univariate models are important in themselves, they also play a key role in providing a “baseline” to which multivariate models may be compared. We shall analyze several multivariate models over the next chapters, but our development begins with the simplest. This is the single-input transfer function-noise model, in which an endogenous, or output, variable $y_t$ is related to a single input, or exogenous, variable $x_t$ through the dynamic model ${ }^1$
$$y_t=v(B) x_t+n_t$$
where the lag polynomial $v(B)=v_0+v_1 B+v_2 B^2+\cdots$ allows $x$ to influence $y$ via a distributed lag: $v(B)$ is often referred to as the transfer function and the coefficients $v_i$ as the impulse response weights.
12.2 It is assumed that both input and output variables are stationary, perhaps after appropriate transformation. The relationship between the two is not, however, deterministic – rather, it will be contaminated by noise captured by the stochastic process $n_t$, which will generally be serially correlated. A crucial assumption made in (12.1) is that $x_t$ and $n_t$ are independent, so that past $x$ ‘s influence future $y$ ‘s but not vice-versa, so ruling out feedback from $y$ to $x$.
12.3 In general, $v(B)$ will be of infinite order and, hence, some restrictions must be placed on the transfer function before empirical modeling of (12.1)becomes feasible. The typical way in which restrictions are imposed is analogous to the approximation of the linear filter representation of a univariate stochastic process by a ratio of low order polynomials in $B$, which leads to the familiar ARMA model (cf. $\S \S \mathbf{3 . 2 5}-\mathbf{3 . 2 7}$ ). More precisely, $v(B)$ may be written as the rational distributed lag
$$v(B)=\frac{\omega(B) B^b}{\delta(B)}$$

# 时间序列分析代考

## 统计代写|时间序列分析代写时间序列分析代考|测试FOR非线性

11.38一个相关的方法考虑了在长记忆过程中测试和建模非线性(例如，Baillie和Kapetanios, 2007)。库普和波特(2001)已经表明，时间序列中不可预测的结构不稳定性也可能产生阈值型非线性的错误证据。Ghysels等人(1996)的一个惊人发现是，在时间序列建模之前通常应用的非线性转换，如X11季节调整过程，也可能导致非线性行为。同样，季节性调整可能会使1999年平稳下来)。最后，正如Van Dijk等人(1999)和De Lima(1997)所讨论的，忽略异常值和非正态也可能导致非线性的虚假证据

## 统计代写|时间序列分析代写Time-Series Analysis代考|传递函数-噪声模型

12.1到目前为止，在本书中开发的模型都是单变量的，因此，时间序列的当前值，线性或其他，只依赖于自身的过去值，也许，一个时间的确定性函数。虽然单变量模型本身很重要，但它们在提供多变量模型可以比较的“基线”方面也起着关键作用。我们将在接下来的章节中分析几个多元模型，但是我们的开发从最简单的开始。这是单输入传递函数-噪声模型，其中内生或输出变量$y_t$通过动态模型${ }^1$
$$y_t=v(B) x_t+n_t$$

12.2假设输入和输出变量都是平稳的，也许经过适当的变换。然而，两者之间的关系不是确定的——相反，它会受到随机过程$n_t$捕获的噪声的污染，该过程通常是串行相关的。(12.1)中做的一个关键假设是$x_t$和$n_t$是独立的，因此过去的$x$会影响未来的$y$，反之亦然，因此排除了从$y$到$x$的反馈。
12.3一般来说，$v(B)$的顺序是无限的，因此，在(12.1)的经验建模变得可行之前，必须对传递函数施加一些限制。施加限制的典型方式类似于$B$中通过低阶多项式的比率逼近单变量随机过程的线性滤波表示，这导致了我们熟悉的ARMA模型(参见$\S \S \mathbf{3 . 2 5}-\mathbf{3 . 2 7}$)。更准确地说，$v(B)$可以写成有理分布滞后
$$v(B)=\frac{\omega(B) B^b}{\delta(B)}$$

## 有限元方法代写

tatistics-lab作为专业的留学生服务机构，多年来已为美国、英国、加拿大、澳洲等留学热门地的学生提供专业的学术服务，包括但不限于Essay代写，Assignment代写，Dissertation代写，Report代写，小组作业代写，Proposal代写，Paper代写，Presentation代写，计算机作业代写，论文修改和润色，网课代做，exam代考等等。写作范围涵盖高中，本科，研究生等海外留学全阶段，辐射金融，经济学，会计学，审计学，管理学等全球99%专业科目。写作团队既有专业英语母语作者，也有海外名校硕博留学生，每位写作老师都拥有过硬的语言能力，专业的学科背景和学术写作经验。我们承诺100%原创，100%专业，100%准时，100%满意。

## MATLAB代写

MATLAB 是一种用于技术计算的高性能语言。它将计算、可视化和编程集成在一个易于使用的环境中，其中问题和解决方案以熟悉的数学符号表示。典型用途包括：数学和计算算法开发建模、仿真和原型制作数据分析、探索和可视化科学和工程图形应用程序开发，包括图形用户界面构建MATLAB 是一个交互式系统，其基本数据元素是一个不需要维度的数组。这使您可以解决许多技术计算问题，尤其是那些具有矩阵和向量公式的问题，而只需用 C 或 Fortran 等标量非交互式语言编写程序所需的时间的一小部分。MATLAB 名称代表矩阵实验室。MATLAB 最初的编写目的是提供对由 LINPACK 和 EISPACK 项目开发的矩阵软件的轻松访问，这两个项目共同代表了矩阵计算软件的最新技术。MATLAB 经过多年的发展，得到了许多用户的投入。在大学环境中，它是数学、工程和科学入门和高级课程的标准教学工具。在工业领域，MATLAB 是高效研究、开发和分析的首选工具。MATLAB 具有一系列称为工具箱的特定于应用程序的解决方案。对于大多数 MATLAB 用户来说非常重要，工具箱允许您学习应用专业技术。工具箱是 MATLAB 函数（M 文件）的综合集合，可扩展 MATLAB 环境以解决特定类别的问题。可用工具箱的领域包括信号处理、控制系统、神经网络、模糊逻辑、小波、仿真等。