Forecasting Islamic securities index using artificial neural networks: performance evaluation of technical indicators

Abstract

PurposeThe purpose of this study is to develop a precise Islamic securities index forecasting model using artificial neural networks (ANNs).Design/methodology/approachThe data of daily closing prices of KMI-30 index span from Aug-2009 to Oct-2019. The data of 2,520 observations are divided into training and test data sets by using the 80:20 ratio, which corresponds to 2016 and 504 observations, respectively. In total, 25 features are used; however, in model selection step, based on maximum accuracy, top ten indicators are selected from several iterations of predictive models.FindingsThe results of feature selection show that top five influencing indicators on Islamic index include Bollinger Bands, Williams Accumulation Distribution, Aroon Oscillator, Directional Movement and Forecast Oscillator while Mesa Sine Wave is the least important. The findings show that the model captures much of the trend and some of the undulations of the original series.Practical implicationsThe findings of this study may have important implications for investment and risk management by using index-based products.Originality/valueNumerous studies proved that traditional econometric techniques face significant challenges in out-of-sample predictability due to model uncertainty and parameter instability. Recent studies show an upsurge of interest in machine learning algorithms to improve the prediction accuracy.

Implementation of ANN and GARCH for Stock Price Forecasting

“…This section technique is based on Aslam et al [5], The creation of a neural network capable of accurately forecasting a financial time series is a difficult problem. The following figures 1 are the various phases in this process.…”

“…They are capable of modeling nonlinear systems without being aware of the relationships between the input and output variables. The nonparametric ANN model may be favored over typical parametric statistical models in cases when the input data do not match the parametric model's assumptions or when the dataset contains big outliers [5]. Second, ANNs are approximations of universal functions.…”

Türkiye’deki İslami Hisse Senedi Endeksinin, Endeks Tabanlı Öznitelikler Kullanılarak Derin Öğrenme Yöntemi ile Tahmini

“…Yavuz, (2019), BIST Katılım 30 ve 50 endekslerini, Markov zinciri analizi ile 0,0065 (artışta), 0,0232 (azalışta) ve 0,0297 (sabitte) mutlak hata ile tahmin ederken, Mar'i, Pratiwi, Oktanisa, & Utaminingrum, (2019) Jakarta İslami endeksini Gauss eliminasyonu, Cramer kuralı ve Gauss-Jordan katsayı belirleme yöntemleriyle sırasıyla, %0,43, %0,44, %0,83 ortalama mutlak yüzdesel hata (MAPE) değeri ile tahmin etmiştir. Aslam et al, (2020), 25 teknik indikatörü kullanarak, Pakistan KMI-30 İslami hisse senedi endeksini 5,29 MAPE değeri ile tahmin etmişlerdir. Bu tahmin yöntemlerinin yanı sıra, literatürde İslami hisse senedi endeksini LSTM ile tahmin eden az sayıda çalışma bulunmaktadır.…”

“…Tahmin modelinin girdi değişkenleri belirlenirken, literatürde yaygın olarak tercih edilen teknik indikatörler (Aslam, Mughal, Ali ve Mohmand, 2020;Doaei, Mirzaei ve Rafigh, 2021) veya makroekonomik göstergeler (Narayan, Phan, Sharma ve Westerlund, 2016;Umam, Ratnasari ve Herianingrum, 2019) kullanılmamıştır. Bunun yerine, endeks tabanlı bir yaklaşım izlenerek, XU100 endeksi (XU100), CBOE oynaklık endeksi (VIX), altın endeksi (GVZ) ve dolar endeksi (DXY) tahmin modelinin girdi değişkenleri olarak belirlenmiştir.…”

Forecasting of Turkish Sovereign Sukuk Prices Using Artificial Neural Network Model

“…There is a massive literature on using ANN for forecasting financial data series such as stock index, exchange rate, gold, and oil prices, etc. (Aslam et al, 2020;Singh & Srivastava, 2017;Tealab et al, 2017;Xu et al, 2020b). On the other hand, there are specific topics in finance where limited studies have been conducted with ANN models.…”