The project titled "Currency Exchange Rate
Prediction'' is the regression problem in Machine
Learning. In the financial market, Current Exchange is
playing the biggest part and expanding its wings day by
day by the concept of Globalization. If you look into the
current US Dollar which speaks 81.40 in Indian rupees.
[fig:1]. Here the value of one US dollar is different from
country to country. There are many factors which affect
the exchange rates of currency like psychological
aspects, political and economic etc. [Fig:3]. The problem
of Currency Exchange prediction is difficult to deal with.
Through this project, our team is going to solve the
problem of currency exchange with Machine learning
technology using python. Predicting the currency rate
gives the investor an added edge in making their
investment in a better method because the forex market
is the foundation of worldwide investing and
international trade. It's crucial to accurately calculate
the forex rate so that we don't give people incorrect
information. EMD-RNN and ARIMA are two models
that we are utilizing to make an accurate prediction. To
demonstrate which is superior, compare their output
with the same data set. The historical dataset obtained
through foreign exchange is used to test the
aforementioned strategies. Predicting currency exchange
rate predictions need to look into all the changes and
consider them daily globally. [Fig:3]. These predictions
affect the income of every citizen of a person and show
impacts on businesses as well as on a country's economy.
Thus, with the currency exchange rate prediction we can
help every individual as well as country in many ways.
The future currency exchange predictions are derived by
studying all possibilities of historical data in the FOREX
Market. There are four Machine Learning models which
support currency exchange predictions. They are
Backpropagation, Radial Basis Function, Long Shortterm Memory, Support Vector Regression.
Keywords :
Currency Exchange Rate Prediction, ARIMA, FOREX Marketing, Regression, Supervised Machine Learning, Decision Node Regression Algorithm, CART.