Claudia
Research question#
-
Influence of coronavirus on European financial markets
-
European financial markets, by performing in-depth analyses and making comparisons among a certain set of European countries.
-
I would like to investigate the variation of market returns in five specific countries being France, Germany, Italy, The Netherlands and Spain.
-
I see very little theory in your motivation: what are the channels according to which the Coronavirus (and geographical and temporal variations in terms of infections and mortality) influence either aggregate stock markets (maybe not so interesting) or individual stock returns?
- You mention: systematic (market) risk and unsystematic (stock) risk
- How can you measure them using a panel regression?
- Mention the literature and explain how other researchers do it.
- You can write an equation to make it specific
- How can you measure these two and compare them before/during coronavirus crisis?
- Asset pricing literature
- z-lib.org
-
My suggestion is that you focus on specific channels, and try to show that these channels actually operate (or not!)
Data
- The data is available and will be easy to find (daily stock returns, firm characteristics, covid data)
- Location in country
- Regional effects
- Location in country
- This is very feasible
Methodology
- There are a lot of confounding factors, e.g. European summit re: coronavirus and financing
- Also: changing regulations, lockdowns
- Have a look at the (available) literature
- The plain coronavirus infection numbers might not be enough to control for all of this
- Try to be precise as possible (splitting up aggregate effects into sub-effects)
Conclusion
- Your research design is feasible, but: you have to think more deeply about the hypotheses, and specifically, about the channels of influence.
- Channels means:
- Impact of Infections (what is the way infections influence stock returns)
- Impact of mortality
- Impact of lockdowns
- Impact of other regulations
- Then, you can formulate clearer hypotheses, and sketch out a methodology (along the same lines as you have right now, i.e. panel data) that would work.
- Channels means: