Model for pannel data with clusters


I'm studying the predictors of the adoption of a methodology in firms with a binary logistic regression with data for 4 years (panel data). The firms that adopt this method Y=1 and 0 otherwise.

The problems that I have are that:

i) My dependent variable is at a firm-level (Y=1 if the firm adopts the method and Y=0 if it does not adopt)
ii) My two study variables are continuous. One is country-level (same value for each of the 8 countries) and the other study variable has values that are different within firms (firm-level).
iii) I have control variables that are country-level predictors (independent variables), industry-level predictors and firm-level predictors. So, it is more likely that the indicators of firms that are nested within industries or countries have a higher probability.
iv) panel data (4 years)
v) Sample size. I have data for 51 firms for the first and second year and 39 and 32 for the second and first year (unbalanced).

I am not sure what model I should use. I was thinking in using a logistic regression with multilevel approach but I'm not sure since I only have 8 industries and 8 countries and I have panel data.

I'd appreciate help regarding the best type of logistic regression model for this situation.
Thank you very much!