Dear Statalist Community,
I am currently working with Stata for the first time and I am encountering some troubles.
I have a panel of 250 entities for the period from 2003 until 2022. I want to analyse the impact of interest rates on some profitability and balance sheet components of those entities, such as net interest income, loans, deposits etc. I decided to work with annual growth rates as I think it better captures the changes. Thus all dependent variables, for example the net interest income are in percentage as growth rate. The independent variable, the interest rate is also expressed in annual changes. I also have a dummy variable for the entities, which is entity size expressed as ln of total assets. Additionally, I wanted to examine if there is a difference between entities with high and low loan-to-deposit ratios (LDR), which is why I have a dummy High-LDR if LDR >85% and Low-LDR if LDR <85%. I also included control variables for GDP growth rate and inflation growth rate. I also created dummy variables for periods where the interest rates were hiking, near zero, or dropping to see these effects. Also, I created an interaction term => interest rate*High-LDR, to see if the effect is stronger with the higher LDR.
Now I am not entirely sure what is the most appropriate approach to examine the effect of changing interest rates on the above mentioned variables (net interest income, loans, etc.). The related literature mostly used fixed effects (Borio et al. 2017) for example.
Should I do some tests of the data before doing the regressions?
I am happy and thankful to get any piece of advice!
Best,
Lukas
I am currently working with Stata for the first time and I am encountering some troubles.
I have a panel of 250 entities for the period from 2003 until 2022. I want to analyse the impact of interest rates on some profitability and balance sheet components of those entities, such as net interest income, loans, deposits etc. I decided to work with annual growth rates as I think it better captures the changes. Thus all dependent variables, for example the net interest income are in percentage as growth rate. The independent variable, the interest rate is also expressed in annual changes. I also have a dummy variable for the entities, which is entity size expressed as ln of total assets. Additionally, I wanted to examine if there is a difference between entities with high and low loan-to-deposit ratios (LDR), which is why I have a dummy High-LDR if LDR >85% and Low-LDR if LDR <85%. I also included control variables for GDP growth rate and inflation growth rate. I also created dummy variables for periods where the interest rates were hiking, near zero, or dropping to see these effects. Also, I created an interaction term => interest rate*High-LDR, to see if the effect is stronger with the higher LDR.
Now I am not entirely sure what is the most appropriate approach to examine the effect of changing interest rates on the above mentioned variables (net interest income, loans, etc.). The related literature mostly used fixed effects (Borio et al. 2017) for example.
Should I do some tests of the data before doing the regressions?
I am happy and thankful to get any piece of advice!
Best,
Lukas
Comment