Please go through this paper it might help.

https://link.springer.com/article/10.1007/s10614-015-9554-z

Use vector error correction modeling (VECM)

The suggestion by @Prince Fosu is good. However, you need to consider some things.

(1) Can you handle the matgematical derivation of these techniques

(2) will your advisor have time or even love to go through your mathematical configuration

(3) Do you have the code or can you write the code to derive these if it is not a ready made approach in any software..

Consider these before making your decision.

If you are going for Multiple regression according to you which I presume to be Ordinary least square, you need to have at the back of your mind that all your variables are level stationary. The suitable approach for statonarity problem in this context is ARDL.

if you are opting for VAR, then there's a mixed feelings that it is either Trade war causes FDI, or FDI causes trade war which has been proved by previous researchers. And also VAR is used when all variable are stationary at level while VECM is used when all variables are I(1).

Best Regards

A VAR model will be a good alternative. The shocks to the model is well captured in VAR modelling and to see at what stage the model settle down after a shock. It is not for forecasting but to trace the impact of the shock using Impulse-response function and variance decomposition.

you can't choose a model without describing the data you have. The choice depends heavily on the nature of data, time-series, panel data... If you have sufficient data, VECM could be a good choice since it gives the trade relationship between American and Chinese economies in the short and long terms. A switching model also may be interesting since it shows the changes in parameters between two states (conditions: stability and war). You can run tests for structural changes.

If you have monthly data it's enough to run these kinds of models. However, if your data is yearly, I doubt you find any model that can be estimated with such few observations.