I have a data set containing 20 years of Gini values for a country. The latest data are for 2018. I want to predict the Gini values for this country by 2025. How can I do this using ML techniques? Also, which econometric forecast model would you recommend?
$\begingroup$
$\endgroup$
1
-
1$\begingroup$ You have a time series, better to look into methods for time series prediction (or forecasting). Search this site! $\endgroup$ – kjetil b halvorsen♦ Feb 23 at 6:40
Add a comment
|
$\begingroup$
$\endgroup$
1
The very first thing you should do is try an extremely simple forecasting benchmark, like the exponential smoothing model that is automatically selected by forecast::ets(). I recommend the excellent free online book Forecasting: Principles and Practice (2nd ed.) by Athanasopoulos & Hyndman.
After you have calculated this benchmark, which should take about ten minutes, you can go into econometric forecasting, putting all other macroeconometric variables into a giant vector autoregression (VAE). There are many textbooks on econometrics. This second step shouldn't take much more than a year.
answered Feb 23 at 6:43
-
$\begingroup$ Thank you Mr. Kolasa. The book was useful to understanding the basics of forecasting. It has also a chapter about how we can forecast with ML techniques. Thank you for your benchmark suggestion, it'll be my first step. $\endgroup$ – Maxpayne Feb 24 at 7:50
