Determinants of Wages Data (CPS 1988)

Cross-section data originating from the March 1988 Current Population Survey by the US Census Bureau.

Usage

data("CPS1988")

Format

A data frame containing 28,155 observations on 7 variables.

wage: Wage (in dollars per week).
education: Number of years of education.
experience: Number of years of potential work experience.
ethnicity: Factor with levels "cauc" and "afam" (African-American).
smsa: Factor. Does the individual reside in a Standard Metropolitan Statistical Area (SMSA)?
region: Factor with levels "northeast", "midwest", "south", "west".
parttime: Factor. Does the individual work part-time?

Details

A sample of men aged 18 to 70 with positive annual income greater than USD 50 in 1992, who are not self-employed nor working without pay. Wages are deflated by the deflator of Personal Consumption Expenditure for 1992.

A problem with CPS data is that it does not provide actual work experience. It is therefore customary to compute experience as age - education - 6 (as was done by Bierens and Ginther, 2001), this may be considered potential experience. As a result, some respondents have negative experience.

Source

Personal web page of Herman J. Bierens.

References

Bierens, H.J., and Ginther, D. (2001). Integrated Conditional Moment Testing of Quantile Regression Models. Empirical Economics, 26, 307–324.

Buchinsky, M. (1998). Recent Advances in Quantile Regression Models: A Practical Guide for Empirical Research. Journal of Human Resources, 33, 88–126.

Examples

#> Loading required namespace: quantreg
## data and packages
library("quantreg")
#> Loading required package: SparseM
#> 
#> Attaching package: ‘SparseM’
#> The following object is masked from ‘package:Matrix’:
#> 
#>     det
#> 
#> Attaching package: ‘quantreg’
#> The following object is masked from ‘package:survival’:
#> 
#>     untangle.specials
data("CPS1988")
CPS1988$region <- relevel(CPS1988$region, ref = "south")

## Model equations: Mincer-type, quartic, Buchinsky-type
mincer <- log(wage) ~ ethnicity + education + experience + I(experience^2)
quart <- log(wage) ~ ethnicity + education + experience + I(experience^2) +
  I(experience^3) + I(experience^4)
buchinsky <- log(wage) ~ ethnicity * (education + experience + parttime) + 
  region*smsa + I(experience^2) + I(education^2) + I(education*experience)

## OLS and LAD fits (for LAD see Bierens and Ginter, Tables 1-3.A.)
mincer_ols <- lm(mincer, data = CPS1988)
quart_ols <- lm(quart, data = CPS1988)
buchinsky_ols <- lm(buchinsky, data = CPS1988)
# \donttest{
quart_lad <- rq(quart, data = CPS1988)
#> Warning: Solution may be nonunique
mincer_lad <- rq(mincer, data = CPS1988)
buchinsky_lad <- rq(buchinsky, data = CPS1988)
#> Warning: Solution may be nonunique
# }