Greetings from my side.
I am working on a panel of Asian countries. However, the biggest hurdle I am facing is missing values/data for financial inclusion. In many papers of financial inclusion, authors have mentioned that they applied factor analysis techniques to create an index for financial inclusion by using several proxies.
I am also trying to use the same proxies by doing "Principle Component Analysis" to create financial inclusion index. Due to missing values, the index does not show constant values throughout the years (some years show blank lines), because some countries do not have data at all for some dimensions. Can I assign the '0' to those missing variables?
Kindly advise me on how to deal with this situation or use any particular technique where I do not have a complete set of data. I shall be very grateful to you.
Best Regards.
Assigning a value of 0 (rather than missing data) will have a strong effect on any variable -- think about what would happen to the mean score. Instead, you should look into the general approach known as "imputation" of missing data.
Dear Sir,
Thank you very much. I am exceedingly obliged.
Best Regards
Imputation of missing data is the correct way of handling missing data. If you insert 0 or any value arbitrarily, it will affect the results of the statistics and subsequently your conclusion.
- Badges
- Science topic
- Similar topics
- Indexes