Econometrics of Time Series 2nd SEMESTER 2021/22 Assignment
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
2nd SEMESTER 2021/22 Assignment
Econometrics of Time Series
Project content
Stock markets are highly volatile in the recent few years due to the Covid- 19 pandemic, trade wars and other geopolitical risk factors. Therefore, it is important to have sufficient time series models for forecasting volatility and risk of stock portfolios. In this project, you will select at least 3 stocks from a given stock market to form an equally weighted portfolio and utilize the dataset in the last five years to build a time series volatility model to predict Value-at-Risk.
Forming groups
Each project group should include 1-4 students while it is possible to submit the project as an individual you can form a group of maximum 4 students. In each group one group member submits the assignment on behalf of all the group members on the learning mall.
Project steps
The steps in the project are for reference only and you are free to follow the steps in your own discretion. You can add more steps depending on the analysis you are going to conduct.
Step 1. Selecting the dataset for your empirical analysis. Choose a sufficient number of stocks that are highly liquid and widely traded in a given stock market. Calculate daily log-returns and conduct preliminary data analysis.
Step 2. Build a time series model for the mean equation and volatility utilizing a suitable ARCH/GARCH model and compare the models you considered. For example, you can utilize the wide range of GARCH models we studies in this course.
Step 3. Use the time series model to generate forecasts of the Value-at-Risk (VaR) of your portfolio at the daily time horizon.
Step 4. Split your dataset as in sample vs. out of sample parts and forecast the volatility and Value-at-Risk in the out-of-sample period. For example, if you have 1000 observations we can split the last 200 for out of sample backtesting. Using an expanding moving window technique you can recalibrate model parameters and produce new VaR forecasts on a daily basis. At the 95% confidence level out of 200 out of sample days your forecasted VaR should be exceeded on approximately 200x0.05 = 10 days if your VaR forecasting is near perfect.
Step 5. Compare the performance of different ARCH/GARCH models you considered in terms of the properties of the residuals obtained, volatility and Value-at-Risk forecasting ability and provide the empirical analysis.
ProjectReport
Theprojectreportshouldincludethefollowingsections.Introduction,LiteratureReview,Data,
Methodology,EmpiricalResults,Conclusion.TheminimumnumberofwordsforthereportIs2,000 wordsandmaximumis4,000words.Idnumbersofall groupmembersshouldbeincludedinthe frontpageofthereportsubmitted.
Assessment
Thefinalmarkwillbebasedontheevaluationofthesubmittedreportaccordingtothefollowing criteria:
(i) Data (20 marks): Your report should have a clearly explained dataset with the indicated time period. The selection of the dataset and specific details such as handling missing observations etc. should be included in the report. Where necessary visual and graphic tools can be used to display the dataset.
(ii) References and literature (20 marks): Referencing within the manuscript and literature search should be complete. References styling must be consistent following the Harvard referencing style. The studies that are utilized should be briefly discussed and referenced.
(iii) Methodology (20 marks): Econometrics methodology that is utilized should be explained clearly and sufficiently. The selection of the specific methodology should be justified given the empirical features observed in the dataset.
(iv) Written report quality and presentation of results (20 marks): Overall, quality of the written report and the clarity of presentation of the results is also important criteria in the marking. Proper use of English grammar and vocabulary is important. Coherent and smooth flow of the report is expected.
(v) R codes (20 marks): All the written R codes that are utilized in the empirical application must be included in the Appendix part. Codes should be clearly put in order according to the use in the research. Whenever needed codes can be referred within the manuscript with the details given in the appendix.
The final mark should also conform to the following criteria:
≥ • Little to criticize • Professional presentation • Copious evidence of full understanding • Independent reading, thinking and working • Contains interesting results or insights • Rigorous Calculations/proofs/arguments support the results
≥ • Outstandingly well-organized • Copious evidence of full understanding • Independent reading, thinking and working • Obtains some interesting results or ideas • Rigorous calculations/proofs/arguments
≥ • Well organized use of most expected materials • Plentiful evidence of clear understanding • Independent thinking and working • Appropriate calculations/proof/arguments • Convincing results
≥ • Generally well organized use of most expected materials • Convincing evidence of understanding • Relevant calculations/proof/arguments • Having some results
≥ • Sensible use of some relevant materials • Some evidence of understanding • Some relevant calculations/proof/arguments
≥ • Includes at least some relevant materials • Limited evidence of understanding
≥ • Materials thin • Poorly organized • Little evidence of understanding
< • Very little materials • Unorganized • With practically no evidence of understanding
2022-04-21