AN APPLICATION OF SEQUENTIAL REGRESSION MULTIPLE IMPUTATION ON PANEL DATA |
| |
Authors: | MICHAEL JOHAN VON MALTITZ ABRAHAM JOHANNES VAN DER MERWE |
| |
Affiliation: | Mathematical Statistics and Actuarial Science, University of the Free State, South Africa |
| |
Abstract: | The relatively new sequential regression multiple imputation (SRMI) method is introduced, with the process of SRMI laid out in detail. The Project for Statistics on Living Standards and Development and the follow‐up KwaZulu‐Natal Income Dynamics Study provide the real panel data on which the methods reviewed are applied. The SRMI process is used to create multiple datasets completed with values imputed for data originally missing, and using the error component model estimation procedures and Rubin's rules, inferences on the panel data are made. Conclusions are drawn as to the applicability of the SRMI process to these data and as to the results of the regression analyses completed. |
| |
Keywords: | C11 C02 C15 C16 D31 C23 D71 Sequential regression multiple imputation SRMI multiple imputation panel data household welfare social capital |
|