Fibonacci retracement pattern recognition for forecasting GBP/USD foreign exchange market
Abstract
This study presents an approach to the Fibonacci retracement implicates a forecast of
future movements in foreign exchange rates (forex) of the previous movement inductive
analysis. The forex market is one of the utmost intricate markets through the
characteristics of high volatility, nonlinearity and irregularity. Meantime, these
characteristics also make it very difficult to forecast forex. The problem are contain
pattern recognition, classification and forecasting. The research objectives are to
recognize the pattern using the Elliott wave pattern, to compare accuracy patterns
classification between K - Nearest Neighbor (KNN) and Linear Discriminant Analysis
(LDA) and to forecast short term forex market using Fibonacci retracement method. The
results show two different type of trend patterns which are uptrend and downtrend. KNearest
Neighbor (KNN) and Linear Discriminant Analysis (LDA) algorithm are the
general pattern recognition method for nonlinearly feature mining from high dimensional
input Elliott wave patterns. Results show that LDA is better than KNN in terms of
classification accuracy data which are 99.43%. Technical analysis by using Fibonacci
retracements for forecasting will be through after the trends of pattern were recognise.
The market trend upward or downward will have a retracement wave before the next
impulse wave approaches new region. Fibonacci price retracements are determined from
a previous low to high swing to identify potential support levels as the market pulls back
from a high. Retracements are also run from a previous high to low swing using the same
ratios, looking for probable resistance levels as the market reverse from a low. After a
significant price movement up or down, the new support and resistance levels are often
at or near these retracement lines. Among of three levels of Fibonacci retracement which
are 38.2%, 50.0% and 61.8% results, the 38.2% shows the best forecasting for Great
Britain Pound pair to US Dollar currency as major pair by using Mean Absolute Error
(MAE), Root Mean Square Error (RMSE) and Pearson Correlation Coefficient (r) as the
statistical measurements which are 0.001884, 0.000019 and 0.992253 for uptrend and
0.001685, 0.000019 and 0.998806 for downtrend. As conclusion, 38.2% is the best
Fibonacci retracement level to forecast forex market for uptrend and downtrend.