Deep Convolution Neural Network with Dropout in Modeling Exchange Rate Volatility

Exchange rate data possesses time-series features such as a trend. Based on convolutional neural network (CNN) deep learning algorithm, which has the advantages of detecting patterns, extracting effective features, finding interdependence time series data, and its computational efficiency, this paper proposes with dropout model-based approach to model forecast exchange rates. In meantime, uses CNN first predict rates corresponding results are compared those CNN-WD. The experimental showed that CNN-WD is superior in terms error value, fitting degree training time. dataset used for research daily period between December 1, 2003, October 15, 2021, comprised 6528 trading observations. Adjusted closing chosen. First, adopts effectively identify patterns extract relevant dataset, making use past 21 days. Dropout regularization then adopted help prevent from overfitting by temporarily removing neuron along all incoming outgoing connections during if generated random value below set rate. This further evaluates reducibility identifiability As an application, next month’s average tea price Kenya.