Seeking Expert Guidance on Global Findex Data

[Indumati Pandey]

Dear Experts,

I would like to seek expert guidance on one methodological issue. Global Findex data are available for 2011, 2014, 2017, 2021 and 2024 rather than annually. Since I am planning to use a panel covering 2011–2024, would it be methodologically acceptable to use linear interpolation to estimate the missing annual observations between the available Findex years?

If interpolation is used, would it be preferable to:

1. interpolate the Findex variables to create an annual balanced panel, or
2. restrict the analysis to the actual Findex survey years and use an unbalanced/less frequent panel?

Thank you in advance for your suggestions.

Best regards

Dr. Indumati Pandey

--