Modeling and prediction using an artificial neural network to study the impact of foreign direct investment on the growth rate / a case study of the State of Qatar

Abstract

Abstract This study came as an attempt to predict the foreign direct investment of the State of Qatar, depending on the model of artificial neural networks and the comparison between its models, because this type of model takes into account the non-linear and stochastic characteristics that characterize the financial and economic chains in general. A multi-layer artificial neural network was built consisting of three layers (the input layer, the hidden layer, the output layer), and the number of training passes was installed 999 times, and the network learning rate was 0.6 and the activation function used is the SIGMOID function using the back propagation algorithm. The MLP (4-10-1) model gave accurate results that are close to the actual values, and it also gave the lowest values for the error measurement criteria represented in the MAE, RMSE and MAPE standards. This reflects the strength of the predicted model, which is consistent with the results of most studies that have been conducted on the subject, both Arab and foreign. It turned out that the feed-forward artificial neural network model is superior to other network models, as the outputs of the hidden layer are inputs for the time following the next time, which can be relied upon as an appropriate method for future prediction of the GDP of the State of Qatar. Also, the forecast values are positive for the period (2020-2040), which encourages increased investor attraction and market recovery in subsequent periods.