Excuse me, but nobody can provide a suitable answer to this question, because too much information is missing.

E.g.: What is the research question? What kind of variables does the dataset include? What is the estimated equation? Could there be endogeneity? etc etc

Nevertheless, I want to mention that your data do not involve many firms and, therefore, your conclusions might suffer from representativity.

Dear Afiqah.

We need more information about what you want to do.

What is the objective of your study before we suggest methods or confirm the method of your analysis.

Regards

You can possibly consider the following;

1. Estimate POLS

2. Estimate Random Effects Model

3. Estimate Fixed Effects Model

Then do Breusch Pagan Lagrangian Multiplier test to choose between option 1 & 2. Hausman test to choose between option 2 & 3.

Since your N