Q2:6 marks
Development economists are often interested in the determinants of production in impoverished parts of the world. An area
of particular focus in recent years has been agricultural output. When farming is successful, there are large positive
spill-over effects for local economies. Conversely when crops fail or livestock die, famine and associated problems such
as the spread of disease often follow. Consequently, by identifying policies that lead to increased agricultural yields,
economic research can contribute to substantial welfare gains.
Your task is to assess the effectiveness of an agricultural assistance program implemented in Tanzania. The assistance
program involved visits to small family farms where farmers were provided with some training and were given access to
fertilizer. Data were taken on farms that did and did not participate in the program, and information was collected twice
– shortly before the program was implemented, and one year afterwards.
The data are available in the workfile Agriculture.wf1. Dummy variables program and after identify farms that
participated, and the time period after the program was implemented. Agricultural output is measured in US dollars per
person-hectare, and is stored in the variable yield. You also have information on the age of the farmer (age), the size
of the farm (size), an index summarizing labour input (labour), rainfall in the last 12 months (rainfall), and an
indicator of whether or not the farmer has completed seven years of schooling (school).
1. To assess the effectiveness of the program a difference-in-differences estimator can be used. Briefly explain the
concept of a “parallel trend” for this estimator and what this assumption implies for the agricultural assistance
program.
2. Why can’t we estimate the effectiveness of the program by simply averaging the output rates of farms that did
and did not participate in the program? Briefly explain.
3. Calculate the difference-in-difference estimate of the effectiveness of the program (in terms of output) based
only upon average rates before and after implementation.
4. Give the regression equation that would be used to calculate the full difference-in-difference estimate employing
the full set of control variables. Also estimate this equation and interpret the results. Why is this estimate preferable
to the estimate given in part 3?
5. Briefly perform some diagnostics on your model. Do the coefficients make sense? Are there any signs of
misspecification? Are the residuals in line with your assumptions?
Q46 marks
Econometricians are often employed as consultants to provide predictions about the likelihood of future events. One of
the most globally important (potentially predictable) events in the next few years will be the US presidential election
in 2016. This will be a contest between nominees from the Democratic and Republican US political parties, and much is
dependent upon the eventual winner.
Imagine that you have been contracted to predict the result of this election. You are required to (i) build a suitable
econometric model, (ii) run some simulations to predict the winner (based upon some inputs which you may assume), and
(iii) write a short justification (2-3 hundred words) of your specific model and its results. You should look to extend
the basic model used in the textbook for this exercise in your own way, and therefore everybody should end up with a
different answer.
Data from previous US presidential elections can be found in the file Fair4.wf1. The variables are defined as follows.
Vote: Vote share of the incumbent party.
Year: Election year.
Party: 1 if the incumbent is a Democrat, -1 if a Republican.
Duration: A measure of the time the current party has held the White House.
War: A dummy for involvement in large foreign wars.
Growth: GDP growth rate in the year of the election.
Inflation: Inflation over the last presidential term.
Goodnews: The number of quarters where growth exceeds 3.2% over the last term.
Person: A dummy variable indicating whether or not the sitting president is running for re-election.
Place this order with us and get 18% discount now! to earn your discount enter this code: special18 If you need assistance chat with us now by clicking the live chat button.