Gold is a valuable metal which is widely used to accumulate capital and also as the raw material for a spectrum of jewellery and technological devices. Considering the broad utilization of gold, it is important to model the gold prices on both the country and the global levels. In this study, the gold price in Türkiye is modelled dependent on various related factors. The import, export and production amounts of gold as well as the exchange rate are considered as the input data affecting the gold prices. The seasonal-trend decompositions of the data are analysed as the first step. Then, a multilayer perceptron type deep learning network is developed in Python programming language for the modelling of the gold prices in Türkiye for the period of 2013M01-2023M02. The 70% of the available data is used as the training data whereas 30% of the data is taken as the test data. The actual gold prices and the results of the developed multilayer perceptron deep learning model are plotted which visually shows that the developed model accurately nowcasts the gold prices. The performance metrics of the developed nowcasting model namely the coefficient of determination, mean absolute error, mean absolute percentage error and the root mean square error are also calculated which further verify the accuracy of the developed model. It is argued that the developed model for the modelling of the gold prices can also be used for other countries or regions.