# ECON 323

ECON 323

Instructions: While cooperating on the assignment is encouraged, plagiarism is not. I
will only accept hand written answers to the interpretation part. Show your work as no
marks will be allocated for the final answer alone.
Use Stata or R to do these. If you choose to use another software, please get my
approval at least a week before the assignment is due.
the following series for January 2015-May 2021:
GDP, chained 2012\$, Statistics Canada. Table 36-10-0434-01 Gross domestic product
(GDP) at basic prices, by industry, monthly (x 1,000,000)
Canada / U.S. Foreign Exchange Rate (EXCAUS) from FRED
Total non resident travellers: Statistics Canada. Table 24-10-0005-01 International
travellers entering or returning to Canada, by province of entry, seasonally adjusted
Note that all series should be for Canada as a whole and on a monthly basis. You
should have 3 series/variables by the time you are done.
(b) What is the difference between chained \$ GDP and constant \$ GDP? This is not
an econometrics question.
(c) If only considering the time periods for which you have data for all variables, what
is T?
(d) What is the mean, median and standard deviation of each series?
(e) Graph all three series. Discuss your graphs. If you do not know what to comment
(f) Should you adjust some of these series for seasonality? Which ones? Why?
(h) Regress the number of travellers on GDP and the exchange rate. Interpret your
coefficients and intercept. Control for seasonality and trends as you see fit.
(i) Why do you think that we are controlling for these regressors? Explain theoretically
the importance of each
(j) Consider that the covid pandemic started in Canada in March 2020. How would
you control for this? Modify your model in (h) taking this into account.
(k) Theoretically speaking (i.e. with your economist/econometrician cap on), how
many lags of the dependent variable do you think would make sense to include as regressors
here? Why? What is (in your opinion) the maximum number of lags that you are willing
to consider here? Why?
(l) Without using trends or deseasonlizing the data, establish the optimal number of
lags of the dependent variable that should be included in this model. Please stop at a
year’s worth of lags regardless of the results you get. What is the optimal number?
(m) Going back to the model you had in (j), let’s consider a model where you will
include lags of the GDP. Try including progressively up to 4 lags of the GDP. Calculate
the long run multiplier for each regression. How many lags do you think is optimal here?
(n) Going back to the model you had in (j), do you think that your errors follow an
AR(1) process?
(o) Do you think that your errors follow an AR(p) process, where p=4?
(p) Assuming that your errors follow an AR(1), run the model in (j) correcting for
this.
(q) Going back to the model you had in (j), do you think that some of the series should
be first differenced? Why or why not?
(r) Run a model where all variables are first differenced and interpret your coefficients.
(s) A big part of being an economist is being able to tell when a model makes sense
and when it doesn’t. I have asked you to run a number of models in this assignment;
which ones do you think worked better than others and why? In light of your results to
questions (a) to (k) what is the optimal model here? Your answer can be a combination
of some of the things you did above. Justify each of your decisions.
(t) What other regressors do you think would have been helpful to include here? Do
you have other suggestions on how to make the model better?