# 金融代写|金融计量经济学代考Financial Econometrics代考|MTH9891 Multiple Time Series Modeling

my-assignmentexpert™金融计量经济学Financial Econometrics代写，免费提交作业要求， 满意后付款，成绩80\%以下全额退款，安全省心无顾虑。专业硕 博写手团队，所有订单可靠准时，保证 100% 原创。my-assignmentexpert™， 最高质量的金融计量经济学Financial Econometrics作业代写，服务覆盖北美、欧洲、澳洲等 国家。 在代写价格方面，考虑到同学们的经济条件，在保障代写质量的前提下，我们为客户提供最合理的价格。 由于统计Statistics作业种类很多，同时其中的大部分作业在字数上都没有具体要求，因此金融计量经济学Financial Econometrics作业代写的价格不固定。通常在经济学专家查看完作业要求之后会给出报价。作业难度和截止日期对价格也有很大的影响。

my-assignmentexpert™ 为您的留学生涯保驾护航 在网课代修方面已经树立了自己的口碑, 保证靠谱, 高质且原创的网课代考服务。我们的专家在金融计量经济学Financial Econometrics代写方面经验极为丰富，各种金融计量经济学Financial Econometrics相关的作业也就用不着 说。

## 金融代写|金融计量经济学代考Financial Econometrics代考|Multiple Time Series Modeling

Multiple time series analysis is concerned with modeling and estimation of dynamic relationships among ‘ $m$ ‘ related time series $y_{1 t}, \ldots, y_{m t}$, based on observations over $T$ equally spaced time points $t=1, \ldots, T$, and also between these series and potential exogenous time series variables $x_{1 t}, \ldots, x_{n t}$, observed over the same time period. We shall explore the use of these techniques in leveraging statistical arbitrage in multiple markets in Chapters $5 \& 6$. We first introduce a general model for multiple time series modeling, but will specialize to vector autoregressive (VAR) models for more detailed investigation. As some concepts are similar to those of the univariate models discussed in Chapter 2 , the presentation will be brief.

Let $Y_{t}=\left(y_{1 t}, \ldots, y_{m t}\right)^{\prime}$ be an $m \times 1$ multiple time series vector of response variables and let $X_{t}=\left(x_{1 t}, \ldots, x_{n t}\right)^{\prime}$ be an $n \times 1$ vector of input or predictor time series variables. Let $\epsilon_{t}$ denote an $m \times 1$ white noise vector of errors, independently distributed over time with $\mathrm{E}\left(\epsilon_{t}\right)=0$ and $\operatorname{Cov}\left(\epsilon_{t}\right)=\Sigma_{\epsilon \epsilon}$, a positive-definite matrix. We consider the multivariate time series model
$$Y_{t}=\sum_{s=0}^{p} C_{s} X_{t-s}+\epsilon_{t},$$
where the $C_{s}$ are $m \times n$ matrices of unknown parameters. In the general setting, the ‘input’ vectors $X_{t}$ could include past (lagged) values of the response series $Y_{t}$. In the context of trading $Y_{t}$ could denote the prices of related assets, $X_{t-s}$ could represent the past values of $Y_{t}$ and the values of volume, volatility, market, industry factors of all related assets, etc. Note that the model (3.21) can be written in the form of a multivariate regression model. An important issue that arises in the multiple series modeling is as follows: Even for moderate values of the dimensions $m$ and $n$, the number of parameters that must be estimated in model (3.21), when no constraints are imposed on the matrices $C_{s}$ can become quite large. Because of the potential complexity of these models, it is often useful to consider dimension reduction procedures such as reduced-rank regression methods described in the last section and more in detail in Reinsel and Velu (1998) [289]. This may also lead to more efficient, interpretable models due to the reduction in the number of unknown parameters. The key quantities that play a role are cross-covariance matrices that relate $X_{t-s}$ to $Y_{t}$, for appropriate values of ‘ $s$ ‘.

## 金融代写|金融计量经济学代考Financial Econometrics代考|Co-Integration, Co-Movement and Commonality in Multiple Time Series

The financial time series generally exhibit non-stationary behavior. For the stationary series, the dynamic relationships among components of the vector time series, $Y_{t}$, has been studied in various ways. While the estimation and inference aspects are similar to the univariate series discussed in Chapter 2 , because of the sheer dimension of $Y_{t}$, there are more challenges in the precise estimation of the model coefficients but there are also more opportunities to study the linkages among the components of $Y_{t}$ that may be insightful. For example, this provides an opportunity to investigate the behavior of a portfolio of stocks. The number of unknown elements in (3.22) is $m^{2}(p+1)$ which can be quite large even for modest values of ‘ $m$ ‘ and ‘ $p$ ‘; therefore considerations of the dimension reduction aspects in modeling (3.22) have been given by Box and Tiao (1977) [54] and Brillinger (1981) [59] among others. The traditional tool used in multivariate analysis for dimension reduction, principal components method is based on the contemporaneous covariance matrix, and thus it ignores the time dependence. While Result 4, stated in the previous section, indicates that the linear combination can exhibit time dependence, we should expect the optimal linear combination to exhibit other desirable features, which are described here.

Identifying the common features among the series have been focused on by several authors; for example, Engle and Kozicki (1993) [127], Vahid and Engle (1993) [318] among others. The series if they are stationary, are studied for their co-movement and if they are non-stationary, they are studied for their co-integration. A method that can capture both aspects can be stated as a reduced-rank autoregressive modeling procedure which we briefly describe below. The basic $\operatorname{VAR}(p)$ model in (3.22) without the constant term can be written as
$$Y_{t}=C Y_{t-1}^{*}+\epsilon_{t},$$

## 金融代写|金融计量经济学代考FINANCIAL ECONOMETRICS代 考|MULTIPLE TIME SERIES MODELING

$$Y_{t}=\sum_{s=0}^{p} C_{s} X_{t-s}+\epsilon_{i},$$

289
. 由于末知参数数量的减少，这也可能导致更有效、可解释的模型。发挥莋用的关键量是相关的交叉协方差矩阵 $X_{t-s}$ 至 $Y_{t}$, 对于适当的值’ $s$ !

## 金融代写|金融计量经济学代考FINANCIAL ECONOMETRICS代 考|CO-INTEGRATION, CO-MOVEMENT AND COMMONALITY IN MULTIPLE TIME SERIES

54

59

127
, 瓦䆖德和恩格尔 1993
318

$$Y_{t}=C Y_{t-1}^{*}+\epsilon_{t}$$

## 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。