Changing from Inflation % from previous year to Change to Inflation Rate

[Shazzle Singh]

Hi,

I managed to gather data for inflation from the IMF but its measure is (Prices, Consumer Price Index, All Items, Corresponding period, previous year, percent).

However, I'm only interested to get the inflation rate for quarter given that my data is quarterly data. To my understanding, I would think that the variable I gathered is in first difference.

This is my variable information

* Example generated by -dataex-. To install: ssc install dataex
clear
input double inf
 7.74278215223094
 10.2464332036316
  12.291933418694
 13.1147540983606
 14.7381242387332
               14
 13.5689851767389
 12.3745819397994
 11.8895966029724
 11.0423116615067
 10.8433734939759
 11.6071428571428
 9.77229601518024
  9.1078066914498
 7.78985507246374
 7.28888888888891
 6.30942091616251
 6.47359454855195
 6.13445378151261
 5.96520298260146
 5.69105691056911
 5.75999999999999
 5.77988915281077
 5.62939796716187
 5.76923076923076
 6.05143721633889
 8.00898203592812
 8.80829015544041
 9.96363636363637
 9.70042796005709
 7.90020790020794
 7.41496598639454
 7.07671957671957
 7.02210663198959
 6.61528580603723
 6.01646611779607
 4.69425571340333
 4.67800729040098
 4.51807228915663
 4.30107526881723
 4.36578171091446
 3.83052814857803
 3.80403458213255
   4.524627720504
 3.90050876201245
 3.80100614868644
 3.49805663520268
 2.57534246575342
 2.33949945593033
  2.4232633279483
 2.30686695278971
 2.24358974358973
 2.60499734183944
 2.41850683491062
 2.14997378080752
 1.98537095088822
  1.2435233160622
 .975359342915802
 1.54004106776183
 1.74180327868853
 2.66120777891503
 2.69445856634468
  2.3255813953488
 2.16515609264855
 .897308075772663
 .990099009900991
 1.38339920948619
 1.77427304090683
 3.06324110671939
 2.69607843137256
 2.29044834307992
 2.22760290556901
 2.15723873441995
  2.2434367541766
 2.28680323963794
 2.32117479867363
 2.25246363209761
 2.47432306255837
 2.04937121564973
 2.68518518518515
 2.89123451124368
 2.91571753986331
 3.42309447740759
 3.11091073038775
 3.52363960749332
 3.93979637007527
  2.5595763459841
  2.0113686051596
  1.0340370529944
 .468483816013621
 1.41996557659209
 2.22888984140592
 4.60554371002131
 2.24671470962272
 1.90920661858297
 1.21593291404613
-1.42682429677945
 .829187396351576
 1.20732722731057
 1.24275062137531
end

1. Is there a code to get the inflation rate for the quarter for the information I've got?
2. Given that this is the first difference of the change in inflation, how do I code such that Stata treats this as a first difference operator? (i.e D.inf)

Thank you.

[Justin Niakamal]

I managed to gather data for inflation from the IMF but its measure is (Prices, Consumer Price Index, All Items, Corresponding period, previous year, percent).

However, I'm only interested to get the inflation rate for quarter given that my data is quarterly data. To my understanding, I would think that the variable I gathered is in first difference.

The inflation rate is the year-over-year change in an underlying price index. So you'll want the change corresponding to the previous year. If you want a quarterly inflation rate, you'll need to make CPI quarterly and then calculate the year-over-year growth. I assume you want inflation for the United States provided from the Bureau of Labor Statistics. Please reference the code below. This will pull directly from the St. Louis Fed's Federal Reserve Economic Data website. If you're not using a Mac, then you can download the data here and insheet the data and run the code. Hope this helps.

local series = "CPIAUCSL" // Consumer Price Index for All Urban Consumers: All Items 
!curl -L https://fred.stlouisfed.org/series/`series'/downloaddata/`series'.csv> '`series'.csv'
insheet using "`series'.csv", names comma clear
erase "`series'.csv"

gen  quarter = qofd(date(date), "YMD") // CPI is a monthly figure 
format quarter %tq!Qq-YY
gen count = 1 

collapse (mean) value (sum) count , by (quarter) // Need to collapse monthly to quarterly 
drop if count != 3 // Drop incomplete quarters 
drop count 


gen inf_rate = round(100*(value/value[_n-4]-1), .1) 
drop in 1/4 // Lost observations from YoY 

tsset quarter, quarterly 
tsline inf_rate

[Shazzle Singh]

Thank you, I understand that. However, I am gathering inflation data from 1980Q1 to 2016Q4 for Norway. As I could not find data on the quarterly inflation rate but rather managed to get the year-over-year change in the inflation rate. Given my data, how do I transform this information to get quarterly inflation rate if it is possible that is?

Thank you.

[Justin Niakamal]

The year-over-year change is the inflation rate. If your data is already quarterly, then no transformation is needed.