5
$\begingroup$

I'm writing my bachelor's thesis this spring, and would like to do it by replicating a modern empirical paper in either macro or finance. I've tried to find a paper by skimming through renowned journals, but most of the publications seem too technical for an undergrad.

One option I’m currently considering is: Mankiw, Romer, and Weil (1992).

Looking for paper suggestions or advice on where to look. Any input is appreciated.

Improve this question
$\endgroup$
2
  • $\begingroup$ By definition a renowned journal does not accept simple stuff! So I would suggest to do something more qualitative or on a dataset that is easy to analyze. $\endgroup$
    – Matteo Bulgarelli
    Commented Jan 11 at 15:26
  • $\begingroup$ Good point! Maybe I should take a look at more applied metrics / macro journals. Surely there are tons of empirical papers than an undergrad is able to replicate. Shouldn't you happen to have any in mind? $\endgroup$
    – elaia
    Commented Jan 11 at 23:17

5 Answers 5

Reset to default
5
$\begingroup$

A paper that should not be difficult to replicate and is about cross-section regressions is Barro's paper titled 'ECONOMIC GROWTH IN A CROSS SECTION OF COUNTRIES' (QJE, 1991)

Improve this answer
$\endgroup$
0
3
$\begingroup$

Any unit-root checking paper will be very easy to replicate even for undergraduate student.

For example; Zivot, E., & Andrews, D. W. K. (2002). Further evidence on the great crash, the oil-price shock, and the unit-root hypothesis. Journal of business & economic statistics, 20(1), 25-44..

Reason why they are easy to replicate is that when papers like this are published they are considered novel because they apply some new unit-root test. Such tests originally have to be painstakingly programmed in some programming language, however over time people and software engineers just start including these tests as parts of pre-programmed packages.

As a result paper that uses test that took weeks to program in 2000s can nowadays be performed by undergraduate in few minutes (you only need basics of R or EViews). Unit-root papers are usually very light on theory and use easy to obtain macro data.

Downside is that papers like this will not impress your advisor, but if you just look for something quick and easy to graduate this or similar paper will do it.

Improve this answer
$\endgroup$
1
  • $\begingroup$ This was an interesting read, thanks for the suggestion! Looking for some challenge and a good grade tho $\endgroup$
    – elaia
    Commented Jan 15 at 22:17
2
$\begingroup$

An interesting, relatively simple task is trying to replicate

Bekaert, Geert & Hodrick, Robert J., 1993. "On biases in the measurement of foreign exchange risk premiums," Journal of International Money and Finance, Elsevier, vol. 12(2), pages 115-138, April.

A working python code, alongside a summary of the description from Fumio Hayashi's book econometrics can be found here. The book will be advanced for undergrad but if you like a bit of a challenge, that might be a good topic.

You would need to have access to Bloomberg, LSEG (formerly Refinitiv) or similar to get reliable quotes. It will also be a bit of a challenge to get the right format for the data and require you to spend some time thinking about foreign exchange conventions.

Edit

Since it seems you have access to Bloomberg, you can run FRD to see the forward rates page. This page has an API (the FX toolkit for excel). I recommend using this API as opposed to the BLPAPI that works in Python and other programming languages because the programmatic API doesn't offer the full breadth of the FX toolkit. You can find a detailed explanation if you run HELP DAPI in the command line and search for bfxforward for example.

Improve this answer
$\endgroup$
1
  • $\begingroup$ The paper seems really interesting at first glance! I work as an RA at my uni's finance department, so I have good access to financial data. I'll take a look at Hayashi next, thanks a lot! $\endgroup$
    – elaia
    Commented Jan 15 at 22:35
2
$\begingroup$

I believe good topic for an undergraduate thesis is to work on the Phillips curve. This topic is not extremely complex, but still offers some challenge. Moreover, it is an evergreen topic in macroeconomics.

A good paper on Phillips curve you could try to replicate is:

Gali, J., Gertler, M., & Lopez-Salido, J. D. (2005). Robustness of the estimates of the hybrid New Keynesian Phillips curve. Journal of Monetary Economics, 52(6), 1107-1118.

I do not think as an undergraduate you need to replicate full paper, just one of their Phillips curve estimation will be enough.

Improve this answer
$\endgroup$
0
$\begingroup$

I think that papers linked to assets role in portfolio may be easy to replicate, as they essentially measure cross-asset correlations. Following Baur & Lucey (2010) positive correlations between two assets means that they are diversifier, negative correlations means that they can act as hedge, and negative correlations during crises means that they are safe-haven assets. So basically you could collect data on financial assets returns (on Yahoo Finance), measure cross-asset correlations between some important assets (such as gold, equity in emerging and developed countries (you can have some on MSCI site) and crypto maybe), and see if the correlations evolved through time ! Depending on your econometrics level, you can potentially do a second-round analysis by explaining the determinants of the evolutions of correlations, or measure the correlations with alternative methods (such as multivariate GARCH models), or measure the impact of the evolution of correlations on portfolio performance.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.