Instruction

The project consists of two research questions. Please use multiple equations models (MEMs) to answer both questions as clearly and completely as possible.  Each question is worth 50 marks, for a total of 100 marks.  This project will constitute 35% of your overall grade in this course.

We suggest that you use R for all empirical work involved. However, you should be able to use another statistical software (e.g. Eviews, Stata, Python, etc.) without a problem. If you do choose to work with an alternative software, please note that support for software-specific issues from the course coordinator and tutors may be very limited.

Please upload your report via the  “Turnitin” submission link  (in the  “Assessment / Research Report 2” folder).   Please note that hard copies will not  be accepted.  At the moment, the due date is 3:59 PM on 13 June 2022, but please check BlackBoard regularly for announcements regarding any changes to this. Your report should be a write-up of your answers (in PDF format, single-spaced, and in 12 font size).1

You are allowed to work on this assignment in groups, i.e., you can discuss how to answer the questions with your classmate(s). However, this is not a group assignment, which means that you must answer all the questions in your own words and submit your report separately. The marking system will check for similarities, and UQ’s student integrity and misconduct policies on plagiarism strictly apply.

Questions

The dataset for this project is contained in report2 .csv, which is an extension of the dataset used in Research Report 1. Recall that in the Research Project 1 dataset, the variables are daily time-series of COVID-19 related measures in the United Kingdom for the period 11

March 2020—07 March 2021. In particular, the dataset contains the following variables:

❼ immobility:  index measuring mobility restrictions, where 0 indicates average (unre-

stricted) immobility and 100 indicates 100% less mobilility than the average;

❼ cases: the natural logarithm of new daily COVID-19 cases;

❼ icu: the natural logarithm of total daily COVID-19 patients in the intensive care unit; ❼ deaths: the natural logarithm of new daily COVID-19 deaths.

The dataset for Research Report 2 is nearly identical to that of Research Report 1—the only difference is that the variable icu contains data for the period 11 March 2020—31 March 2021. That is, icu is extended to include 24 days of observations past 7 March 2021 (which is the end of the sample for all other variables).

Please answer the following two questions using an MEM approach .

1. Use the data provided for the period 11 March 2020—07 March 2021 to forecast ICU occupancy for the period 08 March 2021—31 March 2021. Please describe all potential sources of uncertainty.

Use the observed icu data for the period 08 March 2021—31 March 2021 to qualitatively evaluate and compare forecasts generated by the MEM(s) to the forecasts you generated in Research Report 1. What implications can be drawn from this analysis for further developing / refining icu  forecasting methods?   The break down of marks for this question is as follows (total 50 marks):

❼ forecast model identification (10 marks);

❼ forecasts computation (10 marks);

❼ forecast evaluation and comparison (10 marks);

❼ prediction and interpretation (10 marks);

❼ writing and organisation (10 marks).

2. Use the data provided to compute possible effects of mobility restrictions on new cases, ICU occupancy and deaths. In particular, consider a one-time structural shock to immobility  and analyse the dynamic effects of such a shock on new cases, ICU occupancy and deaths.

In answering these questions, please consider implications for policy, and in doing so, describe the various sources of uncertainty that may these implications.  In addition, please compare the results obtained from MEMs to those obtained from alternative models in Research Project 1. The break down of marks for this question is as follows (total 50 marks):

❼ model identification (10 marks);

❼ estimation and testing (10 marks);