notes Tempo Effects -fertility in US 1917-1980

notes Tempo Effects -fertility in US 1917-1980 - Eco 572:...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Eco 572: Research methods in Demography Fertility in the U.S. 1917-1980 We will use the Heuser cohort fertility tables available in the OPR data archive to have a look at fertility in the U.S. between 1917 and 1980. For a much more extensive analysis see the paper by Schoen (2004) "Timing Effects and the Interpretation of Period Fertility", Demography 41(4):801-819. (Available here if you have access to MUSE.) This handout uses Mata, Stata's matrix programming language, because it simplifies a lot of the calculations. Results are passed back to Stata for plotting. I did some preliminary work to put the archive data in a more usable form. The file heuser.dat is a 36 by 64 matrix of single-year age-period rates for ages 14 to 49 and years 1917 to 1980. This is easy to read into Stata: . infile y1917-y1980 using /// > http://data.princeton.edu/eco572/datasets/heuser.dat (36 observations read) . gen age = 13.5 + _n // ages 14.5 to 49.5 In the sections below we will be looking at period and cohort fertility but it is important to remember that we are dealing with an age-period-cohort surface. In class we will use dynamic 3-D graphs to explore the entire surface. Period Fertility You may want to plot the age-specific fertility rates for selected years, for example 1935, 1945 and 1955 . line y1935 y1945 y1955 age, lp(solid longdash dash) /// > title(U.S. Age-Specific Period Fertility) /// > legend(order(1 "1935" 2 "1945" 3 "1955") ring(0) col(1) pos(1)) . graph export uspasfr.png, replace (file uspasfr.png written in PNG format) The TFRs shows an increase over these years . tabstat y1935 y1945 y1955, stat(sum) http://data.princeton.edu/eco572/heuser.html (1 of 7) [2/12/2008 10:49:21 AM]
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Eco 572: Research methods in Demography stats | y1935 y1945 y1955 ---------+------------------------------ sum | 2.1887 2.4218 3.4983 ---------------------------------------- Let us compute the TFR and the mean age of childbearing for all years, storing the results in new variables called tfr and mac . This requires increasing the number of "observations" to 64. The calculations are best done in Mata using matrix multiplication . set obs 64 obs was 36, now 64 . mata: ------------------------------------------------- mata (type end to exit) ------------------------------------------- : rates = st_data(1::36,1. .64) // we know where rates are : ones = J(36,1,1) : tfr = ones'rates : st_store(.,st_addvar("float","tfr"),tfr') : ages = st_data(1::36,"age") : mac = ages'rates :/ tfr : st_store(.,st_addvar("float","mac"),mac') : end --------------------------------------------------------------------------------------------------------------------- . gen period = 1916 + _n We can now plot the TFR and the mean age of childbearing, which we will do in separate panels . line tfr period, name(tfr,replace) title(Total Fertility Rate) . line mac period, name(mac,replace) title(Mean Age of Childbearing) . graph combine tfr mac, xsize(6) ysize(3) /// > title("U.S. Period Fertility") name(periods)
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/12/2008 for the course ECON 572 taught by Professor Rodriguez during the Spring '06 term at Princeton.

Page1 / 7

notes Tempo Effects -fertility in US 1917-1980 - Eco 572:...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online