I would like to know if it is useful to learn Monte Carlo simulations if I want to focus on macroeconometrics ? Or even is it useful outside of academics ?
$\begingroup$
$\endgroup$
2
-
2$\begingroup$ In short, yes it is. It is becoming increasingly popular to conduct sensitivity analyses for models using Monte Carlo simulations; that would be a good place to look into to start $\endgroup$– BrennanJan 22 '21 at 20:17
-
1$\begingroup$ Very useful indeed. It is one of the most informative ways to illustrate the finite sample performance of your test/estimation procedures, e.g. bias, accuracy, size, power, etc. $\endgroup$– MauOlivaresJan 22 '21 at 23:40
Add a comment
|
$\begingroup$
$\endgroup$
The main challenge in macroeconometrics as opposed to microeconometrics is a the small sample size, because of short recorded time-series and less than 200 countries providing data. To get around this problem increasingly the more popular approach is Bayesian estimation, which often relies on Monte Carlo Markov Chains to construct the posterior distribution of the parameter.
So if you want to understand how the results are obtained, you will need to have a rather deep knowledge of the topic.
