Retrieve data at the following site, which is the university library network's primary site. Go to the university's home page, click on 'library', then on `databases A-Z', choose 'CANSIM multi-dimensional@chass', then on 'main menu Multi-dimensional@chass'. Click on `locate series by series numbers or range'. You then add that series to your cart. You can either extract each series by itself, or place all four of them in your cart. When ready to download, check the interval. The easiest format to use is 'MSExcel Ready'. Then click on 'submit query'. VERIFY THAT YOU HAVE READ IN THE DATA PROPERLY, and that it is properly aligned.
1. CPI monthly, table # 3260022, vector # V41690914, runs until from 1992 to December of 2015, base year 2002
2. CPI annually, table # 3260021, vector # V41693271, runs from 1914 to 2015
3. real GDP, annually, table # 3800017, vector # V3860085, runs from 1961 to 2011 (constant 2002 $, series was terminated in 2011)
4. nominal GDP, annually, table # 3800017, vector # V646937, runs from 1961 to 2011 (series was terminated in 2011)
Note that you do not need the table number; the vector or series number will suffice. You only have to input the vector number ('V' stands for vector).
Tasks a-e consist of dealing with a single times series, namely the CPI.
a) Obtain the monthly CPI (consumer price index) from the CANSIM data base. That is the first series listed above. Use the interval 1992:01 to 2015:12. This 22-year interval includes the base year, but even if it does not, that does not matter. The period-to-period changes are not affected by the choice of base year.
Read the documentation for this series. It should list features such as the base year, the frequency (a monthly series in this case), the interval of availability, and the last time that the series was updated.
b. Calculate the inflation rate between
i) December of 2009 and December of 2008.
ii) December of 2015 and December of 1992 (cumulative over 23 years)
iii) October of 2015 and the corresponding month of base period. (cumulative inflation since the base year of 2002)
These are all easy. You should generate one figure for each part. As a reminder, the equation for the inflation rate between point in time t and an earlier point in time s is:
Inflation rates= ((CPI /CPI,) - 1)*100, for t later than s
These give the cumulative price increases between the two periods (period t and period s)• The base must always be specified. You do not plot anything here. Instead, just show the data series on the spreadsheet, and indicate how you calculated those 3 values.
(Make a new datasheet)
c. Now we move on to annual rather than monthly inflation rates.
Retrieve the second series listed above, which gives the CPI at an annual frequency. These annual figures reflect averages taken '1 12 months of each year. Calculate the annual inflation om year 1972 to year 2015, and then plot this series on a his means that you should generate about 43 figures, Lig the inflation rate between 1971 and 1972, 1972 and 1973, 1973 and 1974, , 2014 and 2015.
Comment a bit on this graph. For instance, in which year was inflation the highest, and in which year was it the lowest? What can you say about overall trends during your period?
d. For this next part, you only generate two numbers. Calculate the average annual inflation rate over the entire period between the year 1990 and the year 2015 in two ways:
i) Take the simple arithmetic average for the annual inflation rates between each pair of years (e.g. 1990-1991,
1991-1992, 1992-1993, 1993-1994, 1994-1995, 1995-1996, 1996-
1997, 1997-1998, 1998-1999, 1999-2000, 2000-2001, 2001-2002,
2002-2003, 2003-2004, 2004-2005, 2005-2006, 2006-2007, 2007-
2008, 2008-2009, 2009-2010, 2010-2011, 2011-2012, 2012-2013, 2013-2014, 2014-2015).
Sum up these year-over-year rates, which should be expressed in percentage terms, and divide by 25. You can take average over a column of figures using a formula in Excel (AVERAGE(firstcell:lastcell))
ii) Take the compounded inflation rate between the starting point and the endpoint of this interval using the following formula. The (1/25) figure means that you raise that expression to the one-twenty-fifth power, which means that you are taking the 25th root of that quantity.
((CPIyear2015/CPIyear1990)**(1/25) - 1)*100
The latter measure is probably more useful, as it gives the annualized growth rate between 1990 and 2014 using the year 1990 as the reference point. The interpretation is that on average prices grew at this constant annual rate relative to price levels in 1990. For part a), the base changed each year, whereas for part b) the base, or reference point, was the same for each year. The figures for parts a) and b) should be slightly different.
a. Calculate the following CPI values. Again, there is no graph, nor is there a data series.
i) Calculate the (would-be) level (not the % change) of the CPI for the year 2015 had we experienced a 5 % annual inflation rate over this entire period from 1990-2015. Just use the formula below, which is the inverse of the formula above. Note that we are taking the 25th power of a quantity. This exercise is called a geometric extrapolation, which is the opposite of interpolation. There is just one number to be produced. CPIyear2015 = CPIyear1990*(1 + 0.05)**25
ii) Calculate the (would-be) value of the deflator in 2014 if we had experienced an annual deflation rate of 1% over this 25-year year period (heaven forbid). This scenario is not that far from reality in the case of Japan. The formula is:
CPIyear2015 = CPIyear1990*(1 - 0. O1)**25
Tasks f-g consist of multi-series analyses that involve the GDP consist of multi-series analyses that involve the GDP of 8 , GDP in real terms, and GDP in nominal terms. We are n to the final two series on the list above. (Make
alLyLucs new datasheet)
f. Now retrieve the two series:
i) Retrieve the nominal GDP and real GDP series.
ii) Take the time series of real GDP denominated in 2002 constant $ and plot it from year 1980 to 2011. Identify the stages of the business cycle. Note that the ordinate of the graph reflects the level, and that the slope of this graph reflects changes in real GDP (but not the % change, which is the growth rate).
iii) Extract the time series of nominal GDP denominated in current dollars and plot it on the same graph over the same interval.
iv) Compare these 2 series and explain the discrepancy between them. They should intersect in 2002. Which series rises more quickly, and why?
q. Do not plot anything in this case.
i) Calculate the values for the implicit deflator by applying the identity: real GDP = nominal GDP / deflator * 100. This should generate another time series.
ii) For each of these three series, calculate the annual % change. This will generate three more time series. Note that you cannot calculate values for 1980 because you would need values for the preceding year of 1979. 1981 is the first year that is applicable.
iii) Next, verify that the percentage change in nominal GDP equals (approximately) the percentage change in real GDP plus the inflation rate (as measured by the GDP deflator). The formula is on page 110 in the textbook. Recall that when we take the natural logarithm of both sides of the following equation, and then take the first difference, or the derivative, of the following formula: real GDP = nominal GDP / deflator * 100, we
obtain: % change in real terms = % change in nominal terms -inflation rate.) You do NOT use this formula; instead, you VERIFY it with these figures. They should only be APPROXIMATELY the same.