Forecasting USD-MNT Exchange Rate — Part 2: Machine Learning

When a good model doesn’t work

Robert Ritz
Dec 3, 2018 · 9 min read

The Problem

The Problem: Given available macroeconomic data, can we accurately forecast USD-MNT exchange rates 3, 6, and/or 12 months in the future?

The Data

Setting It Up

The first and last 5 rows of our initial dataset. 130 rows with 9 columns. The last 4 months have no balance of payments numbers.

Feature Engineering

Create Target Values

The Models

Mean Absolute Error for the models tested. This error calculation was chosen for its interpretability.
Code to generate data and plot below.

Feature Importance

The Forecast

Forecast with 24 month window
Forecast with 48 month window

Conclusion

Mongolian Data Stories

Using data analysis, visualization, and machine learning to tell Mongolia’s story. Interested in writing for Mongolian Data Stories? Send an email to Robert Ritz (editor of MDS) at robertritz@outlook.com.

Robert Ritz

Written by

Data Scientist and Director of LETU Mongolia. Keen observer of Mongolia.

Mongolian Data Stories

Using data analysis, visualization, and machine learning to tell Mongolia’s story. Interested in writing for Mongolian Data Stories? Send an email to Robert Ritz (editor of MDS) at robertritz@outlook.com.