HAND-OUT

January 26, 2012

Writing notes and full empirical papers

A simple instructional guide for Eco7 17 /727D students

Lawrence Dacuycuy, PhD

Economics Department

This document (non ex haustive) merely serves as an instructional guide on how to write short papers or
notes as well as full blown empirical papers (not considered as theses). To achieve optimal learning
ex perience, you are encouraged to read scholarly articles listed in the bibliography.

Notes

1. In writing a note (length: 1 600-2000 words excluding tables and graphs), all that it takes is for
you to read a scholarly article, preferably the short ones that appear in letter type journals (e.g.
Economics Letters, Applied Economics Letters, Short communication sections of some journals).
Here are some important reminders:
a. Be aware of the context. Try to understand the theoretical structure or empirical
strategies employed by the author. If it is an empirical paper, determine whether the
paper establishes causal effects or mere correlation. Papers that highlight correlation are
at best weak for they ignore endogeneity issues that are corrected by causality - based
ev idence. If it is a theoretical paper, zero in on the assumptions, methodology of proof
and results.
b. Be aware that results are quite sensitive to assumptions. In many theoretical studies,
major modifications of assumptions necessarily lead to different results that may have
more insightful implications. For instance, if you can still recall your seminal Solow
model, you know that introducing stochastic technological shocks will further enrich the
model compared to specify ing an entirely deterministic technological shock. Thus,
altering assumptions and investigating how they affect results is one of the better
strategies to come up with a theoretical or empirical note.
c. Another alternativ e is to offer additional empirical ev idence. Some theoretical studies do
not have empirical ev idence. In econometrics, you may come up with a Monte Carlo (or
simulation) experiment to evaluate robustness or prov ide an empirical support to a well
known fact. For instance, Dreher, Kriger and Meierrieks (2011 ) examined whether or not
terrorism has a significant impact on migration.
d. You may also offer a critique to an existing methodology. In Bettin, Lucchetti, and
Zazzaro (2012), the researchers took off by enumerating the structural problems
associated with the macro -approach in examining the link between financial development
and remittances. The claim is that by using microeconometric data, they are in a position
to avoid potential endogeneity problems associated with aggregated data. Additionally,
Zanchi (1998) wrote a note that corrected inferential mistakes in methods that measure
inter-industry wage differentials.

2. Some specific tips

A quick survey of the articles listed in the bibliography reveals the common structure of most notes.
There should be a short introductory section followed by data and model. The next section should
contain your main results and the last section is reserved for your concluding remarks.

1|P a g e
 a. In your introductory section , consider the following:
i. Emphasize what you want to do. It is assumed that you already have a base article
and you should convey the impression that you understood everything about the
base article. Needless to say, you should look for an article that is recently
published to benefit from the creative evolution of ideas.
ii. Compared to the usual structure in writing thesis, short notes usually include
critical literature (only ) in the introductory section. This is to economize on time
and is mainly intended to capture readers attention. You may also consider
answering the following guide questions: Are y ou going to reinforce or v alidate
someones work in a different setting? Are you going to introduce a new
methodology that will bolster the authors findings? Are you going to undertake
robustness checks?
iii. Dont forget to highlight what your study will offer.

b. The section that follows your introduction must contain the design of the study . If y ou are
ex tending a certain econometric model used by the reference author, it is mandatory
that you highlight the relevant equations. This is to inform the reader of the relev ant
target equations. Do not write lengthy literature reviews. Cite literature that is directly
related to your note.

c. Third, report results and provide interpretation thereto. Again, be concise. Please rev iew
the tables that appeared in Dreher, Kriger and Meierrieks (2011).

d. Fourth, write your concluding remarks. This section should highlight your papers
significance.

Relatively long em pirical papers

1. In writing full empirical papers, one needs to have an economic issue that can be validated using a
verifiable set of hy potheses. What does this mean? While econometrics has its own theoretical
foundations, it is still largely seen as the empirical arm of economic theory. Thats why setting the
theoretical framework is critical even in empirical studies. Theory mainly serves as the source of
empirical motiv ation.

2. Suppose you want to investigate sector choice (public versus private sector choice). The theory
stipulates that a person will go to the priv ate sector because the relative utility of doing so far
outweighs that of the public sector. In resolv ing theoretica l concerns, it is imperative that you
offer a theory on wage differentials. In your theoretical section, make sure that your framework
must be able to highlight wage differentials.

3. In your empirical section, you have to determine the appropriate methodology. Recall that the
additiv e random utility interpretation is readily applicable to the task at hand. So you may use
binomial logit/probit because there are only two outcomes.

4. Implement the estimator using software. Be sure to compute for marginal effects and be sure that
you can come up with a good interpretation. Given your knowledge of test procedures, do not
forget to ascertain the validity of functional assumptions that are part of the model. Usually, the
significance of cov ariate is enough to prov ide ev idence that the theory works but you should also
pay attention to the magnitude of the marginal effect. Please dont forget to carry out robustness
checks. Again, check out how this is done in Dreher, Kriger and Meierrieks (2011).

2|P a g e
 5. Interpret the results relative to the theoretical model.

6. In sum, here is the proposed structure

a. Section 1 : Introduction
b. Section 2: Theoretical framework
c. Section 3: Methodological design (Empirical framework)
d. Section 4: Results and interpretation
e. Section 5: Concluding remarks

Summ ary

You may find the following table useful.

Table 1 Some useful tips

Section Dos DONT s

Introduction Clearly identify the Omit the motivational
objective of the basis for the study
note/article and provide Write an excessively
the motiv ation for lengthy introductory
writing the note. section
Clearly highlight the Write an introductory
contribution of the note section that is not
to the literature. coherent with the notes
Provide a concise overall structure
backgrounder
Use a more recent base
article
Methodology Clearly prov ide the Propose a methodology
simulation environment that is difficult to justify .
or methodological design Propose an empirical
Clearly formulate an framework that is largely
empirical framework that inconsistent relative to
is consistent with the objectives.
over-all objective Formulate untestable
Formulate testable hy potheses.
hy potheses Propose a framework
Ensure that replicability that cannot be replicated.
standards are met.
Main results Interpret ex haustively . Offer limited
Undertake all relevant interpretation
pre and post estimation Carry out irrelevant
diagnostics diagnostics.
Carry out all relevant Carry out questionable
hy pothesis tests hy pothesis tests
Carry out robustness
checks.
Present results
excellently.
Concluding rem arks Highlight your papers
contributions to the
literature
Write an excellent
summary

3|P a g e
Bibliography
Bettin, G., Lucchetti, R., & Zazzaro, A. (2012). Financial development and remittances: Micro-
econometric evidence. Economics Letters , 115, 184-186.

Dong, Y. (2010). Endogenous regressor binary choice models without instruments, with an application to
migration. Economics Letters , 107, 33-35.

Dreher, A., Krieger, T., & Meierrieks, D. (2011). Hit and (they will) run: The impact of terrorism on
migration. Economics Letters , 113, 42-46.

Zanchi, L. (1998). Interindustry wage differentials in dummy variable models. Economics Letters , 60,
297-301.

4|P a g e