Feedback Pitching Documents 2025
Giani
You need to be a bit more carefull with the comparison with the Bogmans et al. paper. That paper investigates a specific decision (investment) by a firm. In contrast, you would investigate capital market performance which is subject to all kinds of selection effects. For one thing, like the Bogmans paper points out sensitivity to climate policies leads to less investments which probably will effect market performance. You need to have a stronger argument of why there should be a relation between sentiment and market performance.
Furthermore, you need to think more careful about the timing of the relation. Is sentiment caused by market performance? Has sentiment expected to have an immediate effect on market performance?
I also get the feeling that you are underestimating how many studies have already investigated the effect of different components of the energy transition on market performance of oil and gas firms.
Kaitlyn
From the motivation, it is not entirely clear whether you are interested in the quality of the Centre Point market, the public market, or the total market. In the proposal, you should definitely develop the arguments more about when and why HFT use aggressive versus passive strategies. Does it depend on the structure of the dark pool? The overall market sentiment? Is it different in the public market than in the dark pool? Is this expected to be a short term immediate effect or a long lasting effect?
Rory
The motivation distinguishes between brown and green firms but never really explains the difference. It’s hard to know whether you have a good measure (based on ChatGPT or not) if you do not provide a theoretical definition of what you are trying to measure. My sense is that you are looking at low carbon emission firms for green firms, but there might be high emission mining firms who still benefit from an energy transition such as lithium miners because it is an essential component for batteries.
In terms of the tools, ChatGPT prompting will require programmatic interaction with ChatGPT. I would highly recommend to start trying this out as quickly as possible and add some more description in your proposal.
Lucia
On way to construct the control group could be to look for firms where there is a regular CEO turnover (as in the key paper) that was not forced, i.e. there was an explation.
You will have to develop the link more between the variables (leverage, liquidity, retained earings, …) and the theoretical explanations (distress, risk taking, …)
Meiqi
My first word of advice would be to write out the proposal more in full paragraphs than what you have done for the pitching document. The risk of working largely in bullet points is that the connection between the different parts of the text are lacking. In any research paper it is important to make sure that the link between the motivation, theory, and method are clear and starting from bullet point can disrupt that connection.
The main comment on the research design that I have is that while your setting looks like a difference-in-difference set up. In a way, the treatement comes before the control in that the two types of investors are regulated in the same way after the ban and they are regulated differently before the ban. It would be worthwhile to think a bit more careful what that would mean for the interpetation of the difference-in-difference regression model.