SEM seems one of the options. You would better explain more about the variables; both dependent and independent.

Difference in differences model, treatment effect model. See woodridge econometrics book

To find the effect of a variable on another you could use causality detection or correlation methods.

To cite a few:

For the correlation methods you could use:

- Pearson correlation: to evaluate the linear relationship.

- Spearman correlation: similar to Pearson, but evaluates the monotonic relationship.

- Cross-correlation: similar to Pearson, but it is used to measure the similarity of two variables as a function of the lag of one variable relative to the other.

For the causality methods you could use:

- Granger causality (or G-causality): a variable X g-causes Y if the predictions of the values of Y based on the past of both X and Y are better than the predictions based only on the bast of Y.

- Transfer entropy: a variable X causes Y if the entropy on the prediction of the future values of Y is lower if using the information about the past of both X and Y than the past of Y alone. It is equivalent to Granger causality for Gaussian variables.

- Convergent cross-mapping: based on the theory of dynamical systems, it uses the variables phase space and can be applied to systems where causal variables have synergistic effects.

These are only some of the techniques that one might want to use.

vous pouvez utiliser le modèle VAR ou VCM

Dear Muhammad,

I will suggest you use General Methods of Moment (GMM).

It has been established in the literature that there is a possibility of simultaneity and indegwneity of explanatory variable in the model (Barguellil et al 2018, Alagidede & Muaza 2016, Ojo & Alege 2014).

Regards