125.811 Advanced Risk Analytics
Summer semester, 2024 - 2025
Individual Assignment: Market Risk and Credit Risk Modelling
(30 marks)
Please prepare your modelling report with related tables, figures, interpretations, and explanations based on the following tasks. You need to submit one word or PDF file for your report. Please copy and paste your Eviews result tables into your report.
Part I - Market Risk Modelling (21 marks)
GARCH and EWMA models (Review Week 4 Practice Notes 1)
1. Goto https://www.nasdaq.com and download five-year historical daily closing prices of a stock of your choice.
2. (3 marks) Calculate the logarithm returns for the stock. Analyse and discuss the stock return characteristics, including return mean, median, minimum, maximum, skewness, kurtosis, serial correlation, and volatility clustering.
3. (4 marks) Based on analyses in Step 2, build an appropriate GARCH model for the return volatility and explain why the model specification is appropriate. (Note that: you need to capture the serial correlation of the return level using ARMA model) .
4. (3 marks) Based on the analyses in Step 2, establish an appropriate EWMA return volatility model.
5. (3 marks) Forecast one day volatility using GARCHand EWMA model based on above steps 3 and 4.
Dynamic Conditional Correlation - Bivariate GARCH model (Review Week 4 Practice Notes 2)
1. Using one pair of data (including the spot and futures prices) from the givendata file on stream (file name: Data for assessment 2) . Look at the first tab of the file for the assigned data for your group (students in one group would use the same data, but the assignment is individual assignment) .
2. (4 marks) Estimate their dynamic conditional correlation using the DCC-MGARCH(1,1) model.
3. (4 marks) Based on the estimated dynamic conditional correlation (pt) and estimated standard deviation of the spot (σs,t) and futures (σF,t), calculate the dynamic optimal hedge ratio as follows:
Part II - Credit Risk Modelling (9 marks)
1. Employ the Mortgage.csv and lgd.csv datasets downloaded from
http://www.creditriskanalytics.net/datasets.html
2. (5 marks) Build the best Probability of Default model you could and assess its performance.
3. (4 marks) Find a relevant literature on probability of default model. Explain and discuss the probability of default model in that paper and compare that model with your model (no more than one page of writing) .