#3 - Short
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. Please upload a single file, preferably pdf, to the dropbox on
D2L.
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.
(a) From Statistics Canada’s website, please download the following series from 1961-2016:
1. Long-run provincial and territorial data, Unemployment rate, Annual, Table: 36-10-0229-01
(formerly CANSIM 384-5000)
2. Maintenance and repair expenditures in housing (x 1,000,000), Annual, Total Expenditures,
Table: 34-10-0095-01 (formerly CANSIM 026-0009)
3. Housing Starts, total units, Canada Mortgage and Housing Corporation, under construction
and completions, all areas, annual, Table: 34-10-0126-01 (formerly CANSIM 027-0009)
From the FRED:
3-Month or 90-day Rates and Yields: Interbank Rates for Canada, Percent, Annual, Not Seasonally Adjusted.
Note that all series (other than the FRED one) should be for all 10 provinces individually. You
should have 4 series/variables by the time you are done.
(b) What is N? What is T?
(c) What is the mean and standard deviation of each series for each province separately?
(d) Graph expenditures in maintenance and repairs of housing by province.
(e) Run a pooled OLS regression of the money spent on house repairs/maintenance in a province
on the provincial unemployment rate, housing starts and the 3 month interest rate. Comment on
your results. Are you justified to include all these regressors?
(f) Should you adjust some of these series for seasonality? Why?
(g) Should you adjust your series for trends? Which ones? Why?
(h) Is this a panel or pooled cross-sections? Justify. If it is a panel, do you consider that it is a
micro panel or a macro panel? If this is a panel dataset, is it balanced?