Hi everyone,

I am currently researching in the field of foreign aid, interested what effects certain criteria between donor and recipient have on the level of foreign aid a recipient state is provided. More specifically, over the course of 20 years, I am looking at the effect of

a) A former colonial relationship (Binary = 0;1)

b) A commong language (Binary = 0;1)

c) The volume of annual trade (Floating values)

d) The annual size of a potential Diaspora from the recipient state in the donor country (Floating values)

on

e) the dependent variable, that is the volume of Foreign Aid attributed from donor to recipient in every given year

Up until now, I have only collected some very basic experiences from standard Fixed-Effects models, only including time-variant variables such as trade volume. This is the first time I want to combine both time-constant binary dummy IVs (Colonial tie; Common Language) with IVs that actually change over time (Trade and Diaspora). To my knowledge, this is not possible with standard FE models and I am wondering how to best model my respective data to make causal claims regarding the effect of the respective IVs.

After some research, I am primarily looking at Mixed-Models which would take the form of e.g.:

FOREIGNAIDij = β1xCOLONIALTIE + β2xLANGUAGEij + uiTRADE + ui2DIASPORA + εij

I am wondering if the above-mentioned design would allow to draw inferences regarding the effect of the respective IVs on the levels of foreign aid (disregarding the evident flaws in terms of missing confounders).

If not, I would highly appreciate any hints / advice how to "better" model my respective data, that is which alternative methods (ideally to be applied in R) to use in order to draw conclusions.

Many thanks in advance &

Kind regards