Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: using matrix expression


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: using matrix expression
Date   Wed, 4 Oct 2006 17:32:56 -0400

Sayda--

At 03:42 PM 10/2/2006, Sayda G Wentworth wrote:
>I'd like to convert those nominal earnings to 2003 dollars.
>But the results I get are incorrect. To check why, I try:
>gen realearnings1951=earnings1951/cpi[1,1]
On 10/3/06, David Kantor <kantor.d@att.net> wrote:
Nick Cox has already answered this, but I would add that you have the
choice of loading the CPI  values in a matrix or in data, but you
need to decide that clearly, and then code accordingly.
Nick's answer (redacted) was:
  forval i = 1951/2003 {
      gen realearnings`i' = earnings`i' / cpi[`=`i' - 1950']
  }

But...
I would caution you against copying the cpi into a variable (in the
Editor window?) from Excel--this is dangerous for various reasons, and
does not produce an easy way to deflate nominal dollars, Nick's
solution notwithstanding (think what happens after you accidentally
change the sort order and then try to deflate something).

Instead, make a new dataset with a time variable and the relevant cpi
measures, then merge using the time vars. To do that, you will also
want the earnings data in "long" form (see -help reshape-), which will
help you in other ways, I suspect. (This is SER data, right?)

Another caveat: think about what you want to deflate by (a concern
alluded to by David Kantor).  If you go with the CPI instead of an
earnings index (such as the one used by the US SSA to compute
benefits), at least use the research series where possible. You can
get the CPI-U-RS from http://www.bls.gov/cpi/cpiurstx.htm back to
1978, then get the CPI-U (Series Id: CUUR0000SA0) back to 1913,
inflate the CPI-U by 100/62.1 so that the two series are the same in
Dec 1977, and then merge the CPI-U onto the CPI-U-RS (discarding more
recent values of the CPI-U in favor of the CPI-U-RS), then divide all
by the value of that new variable in 2003.  The output from this
method is pasted below in a format that facilitates usage...

***********Appendix***********
input    yr      cpi_u   cpi_u_rs        cpi      cpi03
   1913   15.94203          .   15.94203   .0595074
   1914   16.10306          .   16.10306   .0601085
   1915   16.26409          .   16.26409   .0607096
   1916   17.55234          .   17.55234   .0655182
   1917   20.61192          .   20.61192   .0769388
   1918   24.31562          .   24.31562   .0907638
   1919   27.85829          .   27.85829   .1039877
   1920   32.20612          .   32.20612    .120217
   1921   28.82448          .   28.82448   .1075942
   1922   27.05314          .   27.05314   .1009822
   1923   27.53623          .   27.53623   .1027855
   1924   27.53623          .   27.53623   .1027855
   1925   28.18036          .   28.18036   .1051898
   1926   28.50241          .   28.50241    .106392
   1927   28.01932          .   28.01932   .1045887
   1928   27.53623          .   27.53623   .1027855
   1929   27.53623          .   27.53623   .1027855
   1930   26.89211          .   26.89211   .1003812
   1931   24.47665          .   24.47665   .0913649
   1932   22.06119          .   22.06119   .0823486
   1933   20.93398          .   20.93398    .078141
   1934    21.5781          .    21.5781   .0805454
   1935   22.06119          .   22.06119   .0823486
   1936   22.38325          .   22.38325   .0835508
   1937   23.18841          .   23.18841   .0865562
   1938   22.70531          .   22.70531   .0847529
   1939   22.38325          .   22.38325   .0835508
   1940   22.54428          .   22.54428   .0841519
   1941    23.6715          .    23.6715   .0883595
   1942   26.24799          .   26.24799   .0979768
   1943   27.85829          .   27.85829   .1039877
   1944   28.34138          .   28.34138   .1057909
   1945   28.98551          .   28.98551   .1081953
   1946   31.40097          .   31.40097   .1172115
   1947   35.90982          .   35.90982   .1340419
   1948   38.80837          .   38.80837   .1448614
   1949   38.32528          .   38.32528   .1430582
   1950   38.80837          .   38.80837   .1448614
   1951   41.86795          .   41.86795    .156282
   1952   42.67311          .   42.67311   .1592875
   1953   42.99517          .   42.99517   .1604896
   1954   43.31723          .   43.31723   .1616918
   1955    43.1562          .    43.1562   .1610907
   1956   43.80032          .   43.80032    .163495
   1957    45.2496          .    45.2496   .1689048
   1958   46.53784          .   46.53784   .1737135
   1959   46.85991          .   46.85991   .1749157
   1960   47.66506          .   47.66506   .1779211
   1961   48.14815          .   48.14815   .1797243
   1962   48.63124          .   48.63124   .1815276
   1963   49.27536          .   49.27536   .1839319
   1964   49.91948          .   49.91948   .1863363
   1965   50.72464          .   50.72464   .1893417
   1966   52.17391          .   52.17391   .1947515
   1967   53.78422          .   53.78422   .2007623
   1968   56.03865          .   56.03865   .2091775
   1969   59.09823          .   59.09823   .2205981
   1970   62.47987          .   62.47987   .2332209
   1971   65.21739          .   65.21739   .2434393
   1972   67.31079          .   67.31079   .2512534
   1973   71.49758          .   71.49758   .2668816
   1974   79.38808          .   79.38808   .2963348
   1975   86.63446          .   86.63446   .3233836
   1976   91.62641          .   91.62641   .3420172
   1977   97.58454          .   97.58454   .3642573
   1978    104.992      104.3      104.3   .3893244
   1979   116.9082      114.1      114.1   .4259052
   1980   132.6892      126.7      126.7   .4729377
   1981   146.3768      138.6      138.6   .5173572
   1982   155.3945      146.8      146.8   .5479657
   1983   160.3865      152.9      152.9   .5707353
   1984   167.3108        159        159    .593505
   1985   173.2689      164.3      164.3   .6132886
   1986   176.4895      167.3      167.3   .6244868
   1987   182.9308        173        173   .6457633
   1988   190.4992      179.3      179.3   .6692796
   1989   199.6779        187        187   .6980217
   1990    210.467      196.3      196.3   .7327361
   1991   219.3237      203.4      203.4   .7592385
   1992   225.9259      208.5      208.5   .7782755
   1993   232.6892      213.7      213.7   .7976857
   1994   238.6473      218.2      218.2    .814483
   1995   245.4106      223.5      223.5   .8342665
   1996    252.657      229.5      229.5   .8566629
   1997   258.4541      234.4      234.4   .8749533
   1998   262.4799      237.7      237.7   .8872714
   1999    268.277      242.7      242.7    .905935
   2000   277.2947      250.8      250.8   .9361702
   2001   285.1852      257.8      257.8   .9622993
   2002    289.694      261.9      261.9   .9776036
   2003   296.2963      267.9      267.9          1
   2004   304.1868      275.1      275.1   1.026876
   2005   314.4928      284.3      284.3   1.061217
end
keep yr cpi03
save /cpi03
use yourdata
gen n=_n
reshape long earnings, i(n) j(yr)
sort yr
merge yr using /cpi03
gen rearn=earn/cpi03
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index