统计代写|金融统计代写financial statistics代考|Bubbles and crises

statistics-lab™ 为您的留学生涯保驾护航 在代写金融统计financial statistics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。我们的专家在代写金融统计financial statistics代写方面经验极为丰富，各种代写金融统计financial statistics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

统计代写|金融统计代写financial statistics代考|The Augmented Dickey–Fuller test

It is well known in the unit root literature that the limit distribution of the ADF statistic depends on both the null hypothesis and the precise regression model specification. ${ }^{c}$ Appropriate choices of both therefore have a material impact in practical implementation.

The null hypothesis $\left(H_{0}\right)$ of the PSY test captures normal market behaviors and states that asset prices follow a martingale process with a mild drift function such that (Phillips et al., 2014)
$$y_{t}=g_{T}+y_{t-1}+u_{t},$$
where $g_{T}=k T^{-\gamma}$ (with constant $k, \gamma>1 / 2$, and sample size $T$ ) captures any mild drift that may be present in prices but which is of smaller order than the martingale component and is therefore asymptotically negligible.
The regression model chosen for the PSY procedure is
$$\Delta y_{t}=\mu+\rho y_{t-1}+\sum_{j=1}^{p} \phi_{j} \Delta y_{t-j}+v_{t},$$

where for implementation purposes the regression error $v_{t}$ is assumed to satisfy $v_{t}^{i . i . d} \sim\left(0, \sigma^{2}\right)$. The $p$ lag terms of $\Delta y_{t}$ are included to take care of potential serial correlation. The lag order $p$ is often selected by information criteria. The regression model includes an intercept but no time trend and nests the null hypothesis as a special case with $\mu=g_{T}$ and $\rho=0$. The ADF statistic is simply the $t$-ratio of the least squares estimate of the coefficient $\rho$.

The i.i.d error condition may be replaced with a martingale difference sequence (mds) condition in (2). More general specifications on the error $u_{t}$ in the generating mechanism (1), such as those in Assumption 1 below, may be employed and are accommodated by allowing the regression lag order $p \rightarrow \infty$ as $T \rightarrow \infty$ in (2). Nonparametric adjustments for serial correlation may also be used, such as those developed in Phillips (1987) and Phillips and Perron (1988).

统计代写|金融统计代写financial statistics代考|The Recursive Evolving Algorithm

The recursive evolving algorithm of PSY enables real-time identification of bubbles and crises while allowing for the presence of multiple structural breaks within the sample period. Phillips et al. (2015a,b) show that this algorithm is superior to the forward expanding and rolling window algorithms in bubble identification, especially when the sample period contains multiple bubbles.

For the convenience of exposition, we use the standard “fraction of the total sample” notation for observations. Thus if $t=\lfloor T r\rfloor$ is the integer part of $T r$, then observation $t$ is represented fractionally as observation $r$ and then the total sample runs over values of $r$ from 0 to 1 . Suppose the observation of interest is $r^{\dagger}$. The PSY procedure calculates the ADF statistic recursively from a backward expanding sample sequence. Let $r_{1}$ and $r_{2}$ be the start and end points of the regression sample. The ADF statistic calculated from this sample is denoted by $A D F_{r_{1}}^{r_{2}}$. We fix the end point of all samples on the observation of interest such that $r_{2}=r^{\dagger}$ and allow the start point $r_{1}$ to vary within its feasible range, i.e., $\left[0, r^{\dagger}-r_{0}\right]$, where $r_{0}$ is the minimum window required to initiate the regression. The recommended setting of $r_{0}$ for practical implementation is $r_{0}=0.01$ $+1.8 / \sqrt{T}$. The PSY statistic is the supremum taken over the values of all the ADF statistics in the entire recursion, which is represented mathematically as
$$P S Y_{r^{\dagger}}\left(r_{0}\right)=\sup {r{1} \in\left[0, r^{\dagger}-r_{0}\right], r_{2}=r^{\dagger}}\left{A D F_{r_{1}}^{r_{2}}\right} .$$
The supremum enables the selection of the “optimal” starting point of the regression in the sense of providing the largest ADF statistic.

The PSY test can be conducted for each individual observation of interest ranging from $r_{0}$ to 1 , i.e., for $r^{\dagger} \in\left[r_{0}, 1\right]$. The recursive calculation evolves as the observation of interest moves forward and therefore the procedure is called a recursive evolving algorithm. See Fig. 1 for a graphical illustration of the algorithm. The corresponding PSY statistic sequence is $\left{P S Y_{r^{+}}(r 0)\right}_{r^{*} \in[r 0,1]}$.

统计代写|金融统计代写financial statistics代考|The Rationale

To illustrate the idea of bubble identification, consider the present value asset price formula
$$P_{t}=\sum_{i=0}^{\infty}\left(\frac{1}{1+r_{f}}\right)^{i} \mathbb{E}{t}\left(D{t+i}\right)+B_{t},$$
where $P_{t}$ is the price of the asset, $D_{t}$ is the payoff received from the asset, $r_{f}$ is the risk-free interest rate, $\mathbb{E}{t}(\cdot)$ is the conditional expectation operator given information to time $t$, and $B{t}$ is the bubble component. The bubble component satisfies the submartingale property (Diba and Grossman, 1988)
$$\mathbb{E}{t}\left(B{t+1}\right)=\left(1+r_{f}\right) B_{t} .$$
In the absence of a bubble, the degree of nonstationarity of the asset price is controlled entirely by the dividend series and hence is believed from empirical evidence to be at most $I(1)$. On the other hand, asset prices will be explosive in the presence of a bubble component in formula (7) whenever the initialization $B_{0}>0$ in (8).

Asset price dynamics over the expansionary phase of a bubble period may be modeled in terms of a mildly explosive process (Phillips et al., 2011; Phillips and Magdalinos, 2007; Phillips and Yu, 2009) of the form
$$\log P_{t}=\delta_{T} \log P_{t-1}+u_{t},$$
where the autoregressive coefficient $\delta_{T}=1+c T^{-\eta}$ mildly exceeds unity (with $c>0$ and $\eta \in(0,1))$ and yet still lies in its general vicinity. Detection of a bubble process in the data is therefore equivalent to distinguishing a martingale process of asset prices from a mildly explosive process. This can be achieved by the PSY procedure with null and alternative hypotheses specified as
\begin{aligned} &H_{0}: \mu=g_{T} \text { and } \rho=0, \ &H_{A}: \mu=0 \text { and } \rho>0 . \end{aligned}

统计代写|金融统计代写financial statistics代考|The Augmented Dickey–Fuller test

Δ是吨=μ+ρ是吨−1+∑j=1pφjΔ是吨−j+在吨,

iid 错误条件可以替换为 (2) 中的鞅差序列 (mds) 条件。有关错误的更一般规范在吨在生成机制 (1) 中，例如下面的假设 1 中的那些，可以通过允许回归滞后顺序来使用和适应p→∞作为吨→∞在 (2) 中。也可以使用序列相关的非参数调整，例如 Phillips (1987) 和 Phillips and Perron (1988) 开发的那些。

统计代写|金融统计代写financial statistics代考|The Recursive Evolving Algorithm

PSY 的递归演化算法能够实时识别泡沫和危机，同时允许在样本期间存在多个结构中断。菲利普斯等人。（2015a，b）表明该算法在气泡识别方面优于前向扩展和滚动窗口算法，尤其是在样本周期包含多个气泡时。

P S Y_{r^{\dagger}}\left(r_{0}\right)=\sup {r{1} \in\left[0, r^{\dagger}-r_{0}\right], r_{2}=r^{\dagger}}\left{A D F_{r_{1}}^{r_{2}}\right} 。P S Y_{r^{\dagger}}\left(r_{0}\right)=\sup {r{1} \in\left[0, r^{\dagger}-r_{0}\right], r_{2}=r^{\dagger}}\left{A D F_{r_{1}}^{r_{2}}\right} 。

PSY 测试可以针对每个感兴趣的观察进行，范围从r0为 1 ，即，对于r†∈[r0,1]. 递归计算随着感兴趣的观察向前移动而进化，因此该过程称为递归进化算法。有关该算法的图解说明，请参见图 1。对应的 PSY 统计序列为\left{P S Y_{r^{+}}(r 0)\right}_{r^{*} \in[r 0,1]}\left{P S Y_{r^{+}}(r 0)\right}_{r^{*} \in[r 0,1]}.

统计代写|金融统计代写financial statistics代考|The Rationale

H0:μ=G吨 和 ρ=0, H一种:μ=0 和 ρ>0.

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。