It depends on the kind of data you are dealing with. If time series are involved, than Joao Paulo's answer is correct. If you are instead dealing with cross sectional data (particularly if they refer to different countries, regions, and so on), you should be aware of possible indirect causal links. FDI, in other words, might be regarded as a proxy of the country/region attractiveness, which, in its turn, depends on a number of factors. Therefore, we should be very careful about inferences on the direct effects of FDI.

See, for instance, https://media.law.wisc.edu/s/c_360/tcyzd/foreign_investment_recommended1.pdf