Alpaslan Akay, "Finite Sample Comparison of Alternative Methods for
Estimating Dynamic Panel Data Models", Journal of Applied Econometrics,
Vol. 27, No. 7, 2012, pp. 1189-1204.

The dataset used in the empirical illustrations of the paper is called 
Longitudinal INdividual DAta (LINDA), which is a register-based dataset 
for Sweden. The dataset is built using different sources such as LNU 
(The Level of Livings Survey) and HUS (The Household Market and 
Non-market Activities). The core registers consist of the income 
registers and population censuses.

LINDA is a joint endeavor between the Department of Economics at Uppsala 
University, The National Social Insurance Board (RFV), Statistics 
Sweden, and the Ministry of Finance of Sweden. LINDA is confidential and 
cannot be published online. Permission to access the data can be 
obtained through Statistics Sweden (www.scb.se) subject to the normal 
confidentiality provisions. For more information about the dataset, 
please see the working paper by

  Per-Anders Edin and Peter Fredriksson (2000). "LINDA-Longitudinal 
  INdividual DAta for Sweden", Working Paper Series 2000:19, Uppsala 
  University, Department of Economics.

The part of LINDA used in this paper covers the period 1992 and 2001. 
The sample selection criteria is continuously married women between 
years 1992 and 2001 with positive husband's annual earnings and hours 
worked each year. The dataset also contains information about the 
individual id (bidnr) to be able to identify the individuals to exploit 
the panel dimension. I use the variables: age of the individual, place 
of birth, highest level of education, number of children, earnings of 
husband and yearly working hours.

Alpaslan Akay
akay [AT] iza.org