Hi every one!
I hope this message finds you well. I am currently engaged in analyzing data extracted from periodic household surveys, which have been conducted at irregular intervals, (typically approximately every 7 years). I am particularly interested in understanding the best approach for addressing missing data in this context. Specifically, I would like to inquire whether multiple imputation or interpolation techniques would be more suitable for handling missing data in periodic household survey datasets.
Your expertise would be invaluable in helping me ensure the accuracy and reliability of my analysis results. Thank you in advance for your time and assistance.
Warm regards,
Imputation methods include mean or median imputation, regression imputation, hot deck imputation, and time series imputation. Mean or median imputation fills missing values with the mean or median of observed values, while regression imputation uses other variables to predict missing values. Hot deck imputation matches cases with missing data to similar cases with observed data.
Time series imputation uses data from previous time points to fill in missing values for surveys conducted over time.
Hi Sara, maybe have a look at those papers:
https:\\ro.uow.edu.au/cgi/viewcontent.cgi?article=1094&context=cssmwp
https:\\citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=9a7cddb7f8d4fcb13b192cc806d9c0303f3e
Kind regards, Volker
The single imputation methods mentioned by Behailu Dessalegn all have serious problems and should be avoided. What is currently considered state of the art is multiple imputation and full information maximum likelihood estimation. See
Enders, C. K. (2022). Applied missing data analysis (2nd ed.). Guilford Press.
Enders, C. K. (2023). Missing data: An update on the state of the art. Psychological Methods. Advance online publication. https://doi.org/10.1037/met0000563
Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7, 147-177.
- Badges
- Science topic