# 本次Statistics**统计学代写**的主要内容是使用Stata完成一个时间序列的Project，这是一个开放性题目，需要自己从网上下载Data Set然后将完整的分析过程以report的形式呈现。

**要评估的学习成果是：**

1.选择并应用适当和相关的计量经济学和统计技术，方法和技能；

2.对有效收集，解释，分析，整理和显示数据的特定方法的相对重要性做出可靠的决定；

3.应用计量经济学技术研究经济理论中的问题。

**1.提交指导**

•创建“时间序列项目”文件夹：任何包括数据集的工作，Stata工作和最终报告（Word格式）都必须包括在内。所有学生必须在2020年8月9日（星期日）23:59:00之前将您的“计量经济学项目”文件夹通过电子邮件发送至

•所有学生必须在2020年8月9日星期日23:59:00之前通过Turnitin提交最终报告（Word格式）。提交的内容确认该作品是学生自己的作品。

•Turnitin链接将打开五天，直到2020年8月14日下午6点才可以提交。请阅读MOLE中的延迟罚款政策。

•提交的报告应进行校对，以便进行拼写和语法检查。为页面编号，为部分编号。

•为减少Turnitin技术错误的风险，建议所有学生至少在截止日期前半天提交您的最终作业。

•任何基于计算机的技术问题（例如，计算机宕机），学生犯的任何提交错误（例如，提交错误的版本）都不会被视为重复申请或特殊情况下受益的理由。

**2.主题**

这些说明提供了有关如何利用计量经济学课程和Stata Workshops中学习到的技能进行应用计量经济学项目的建议。

每个学生都必须参加一个单独的应用时间序列计量经济学项目。 广泛的主题是可以接受的。 它必须涉及一些时序经济计量估计。 每个学生都可以选择自己的项目主题，但是研究应该与经济学或商业领域相关，并且需要对数据进行分析。

该项目的重点可能会有所不同。 例如，有关某个主题的现有工作的重要摘要，然后是您自己的计量经济学结果。 或者，以其他地方的研究为基础的报告也是可以接受的。 但是您需要将它们的结果与您的结果进行比较，并尝试解释差异。 您的数据可以来自其他国家或涵盖不同的时间段。 您甚至可以使用相同的数据，但是使用不同的计量经济学方法。

**3.资料**

找到合适的日期可能是项目中最困难的部分。 在决定任何主题之前，应检查数据是否可用，并确保在开始使用数据集之前已对数据集进行了彻底检查。

您需要确保数据大小足以进行分析。 样本量对于确定可以使用的技术和结果的准确性很重要。 建议该学生对年度数据至少进行30次观察，对季度或每月数据进行更多观察。 除非您有处理大型数据集的经验，否则不要超过100个观察值。

The learning outcomes to be assessed are:

1. select and apply appropriate and relevant econometric and statistical techniques, methods

and skills;

2. make robust decisions about the relative importance of particular methods of collecting, interpreting, analysing, collating and displaying data effectively;

3. apply econometric techniques to investigate issues in economic theory.

1. Submission Instruction to students

- Create the “Time Series Project ” folder: any work includes data set, Stata work and final report (Words format) must be included. All students MUST email your “Econometrics Project” by 23:59:00 Sunday 9th of August 2020.
- All students MUST submit final report (Word format) through Turnitin by 23:59:00 Sunday 9th of August 2020. The submission confirms that the work is student’s own work.
- The Turnitin link will open for five days until 6PM 14th August 2020 for late submission. Please read late penalty policy in MOLE.
- Submitted report should be proofread for spelling and grammar. Number the pages, and number the sections.
- To reduce the risk of Turnitin technical errors, all students are suggested to submit your final work at least half day before the deadline.
- Any computer-based technical issues (e.g. compute crash), any submission errors made by students (e.g. submit wrong version) will not be considered as the rationale for benefit of double or Special Circumstances application.

2. Topic

These notes provide advice on how to undertake the applied econometrics project using the skills learnt in the Econometrics course and Stata Workshops.

Each student is required to undertake an individual applied Time-Series Econometrics Project. A wide range of topics is acceptable. It must involve some time-series econometric estimation. Each student is to choose his or her own project topic, but the research should be related to the field of economics or business and requires the analysis of data.

The emphasis of the project can vary. For example, a critical summary of the existing work on a topic, followed by your own econometric results. Or, a report based closely on research reported elsewhere is acceptable. But you need to compare their results with yours and try to explain the difference. Your data can be from a different country, or cover a different time period. You might even use the same data, but use a different econometric approach.

3. Data

Finding the appropriate date can be the most difficult part of the project. You should check that the data is available before deciding on any topic, and make sure you have thoroughly examined your dataset before you start using it.

You need to make sure the data size is big enough for your analysis. The sample size can be important in determining the techniques you can use and the precision of your results. The student is suggested to have at least 30 observations for annual data, and more for quarterly or monthly data. Unless you have the experience of dealing with large data sets, do NOT go much above 100 observations.

You need to understand the definition of your data and the sort of questions you should consider. For example:

- Are the data current or constant price level?
- Are they seasonally adjusted already?
- Has the definition of the data changed over the sample period?
- If it is constructed data, how was this done? You may have to do some work to make the data useable for the project. If so, please provide information on this work in your appendix. So you can be given credit. You may have to adjust the data in various ways to deal with missing observations. Published data are not infallible, so you are always required to double check for possible mistakes. You should know the information about the relevant history of the selected data set, such as important events (strikes, wars), or changes of government.

Once you have downloaded the data onto the computer, you should conduct a descriptive analysis. You could plot the data and note the distinctive features such as trends, seasonal, cyclical behaviour. You should repeat this process after any transformation of the data (ratios, logarithmic transformations).

You should keep detailed notes on the sources of data, and provide reference on your selected data.

Your raw data MUST be provided in the Appendix. When you print out your data set, you should check carefully for typing errors.

You should always keep at least two backed up copies of your data, Stata output, and your report on separate (hard) drive, stored separately. There are many ways of losing or corrupting drives and it can be a lot of work typing any of work again.

4. Suggested Sources

- Economic Trends Annual Supplement
- The Blue Book (National Income and Expenditure)
- The Annual Abstract of Statistics
- Financial Statistics
- Employment Gazette
- Family Expenditure Survey
- Census of Production
- World Development Report
- OECD Main Economic Indicators
- International Financial Statistics
- Economic Report of the President
- World Development Indicators (from World Bank Online database)

5. Format of your Report

You are advised to leave plenty of time for writing up. You can get an idea of the usual format from browsing through economic journals such as The Review of Economics and Statistics. The report is divided into sections, often something like this:

[10 marks] Section 1: Introduction

You research question is stated in words. The motivation is given here for your own contribution. At the end of this section, briefly summarize the contents of the rest of the report, section by section.

[20 marks] Section 2: Description of the dataset

Inform sources of the data, describe the data, definition of variables, any possible measurement errors, how they were collected, etc. When working with time series variables measured in currency units, make sure the units are consistent across variables unless there is a special reason to do otherwise.

It is good to describe the distinctive features of the data, e.g. plots or tables of time series, means, standard deviation, mins and maxes of variables. In addition, the students need to exam whether or not the data is stationary in levels or in differences. You might consider transforming the data format (ratios, logarithmic, difference) for regression modelling.

[30 marks] Section 3: Regression Model

You should use the probabilistic structure of the data to choose an appropriate regression model. The regression model is presented here in more detail, both in words and in mathematical form if appropriate. You need to explain why each independent variable is used. You might discuss your choice of regression model in terms of the assumptions it involves. You need to convince the reader that you have made an appropriate choice. In addition, the regression model, along with any estimation and model specification issues are discussed.

[20 marks] Section 4: Results: Estimation, Interpretation and Forecasting

- Coefficient estimates, specification tests, etc., are presented and interpreted in light of the research question mentioned in the Introduction.
- You might find the following explanation/knowledge is useful to interpret the results:

- ✓ A series displays autoregressive (AR) behaviour if it apparently feels a “restoring force” that ends to pull the series back to its mean. For example, In a AR(1) Model, the AR(1) coefficient determines how fast the series tends to return to its mean. More general, AR(p), where p>1, the sum of the coefficients determines the speed of mean reversion.
- ✓ A series displays moving-average (MA) behaviour if it apparently undergoes random “shocks” whose effects are felt in two or more consecutive periods. The MA(1) coefficient is (minus) the fraction of last period’s shock that is still felt in the current period. The MA(2) coefficient is (minus) the fraction of the shock two periods ago that is still felt in the current period, and so on.
- Try to make sense of the results. Avoid simply reporting the numerical results in the text.
- Stata econometric strategy, estimation method, and software was used
- Give precise variable descriptions for all variables used
- An example about how to write down your estimated regression model (estimated ARIMA (p,0,q) for ?):

• Provide the forecasted/predicted values for your dependent variable, and graphs if it is possible.

[5 marks] Section 5: Conclusion

You could explain the significance of the results, and how they relate to the research question/topic.

[5 marks] Section 6: Reference

A list of the sources that you have referred to in the report must be provided, ordered alphabetically by the first author’s last name.

[10 marks]Section 7: Tables, Figures and Appendices

The data, Stata output must be provided in this section. If there is any Table, Figure, they can be put in the Appendix. Appendices are handy if you have some background information or results that might be of interest to some readers, but are not central enough to belong in the main part of the report.