Quantitative Neural Network Models

seekers_:
Trading activities are based on technical analysis, market sentiment (asymmetric information, rumours, noise trading) and imitative behavoiur. This leads to unjustified biasness in decision making. To remove such subjectivity, this paper suggests a neural network model for the investors to decide whether buy or sell the shares. The model consists two wings - one, based on technical analysis and the other, on fundamental analysis. The integral part of this model is the existence of a hidden layer between the input layer and output layer. To remain away from the subjectivity, this model does not consider the behavioural factors in modeling.

According to studies on the impossible trinity, under conditions of high financial integration, the domestic interest rate is closely linked to the foreign one if the possibility of maneuvering interest rates is absent in this transaction. The Fisher effect is brought into this escapade because interest rates generally trend positively with inflation. Botswana has set her inflation target between 3-6% and this study attempts to determine inflation spillover effects from the United Kingdom, United States of America, Canada, Japan, China, Belgium, France, Germany, South Africa, Nigeria, and Ghana using data from 1980-2012. Comparatively, the attempts made by previous studies to examine spillovers generally lacked a long-run focus and channeled much attention to periods of financial crisis. This study deviates from other studies by using the Augmented Dickey Fuller (ADF) test to examine unit roots for the countries under examination. The study further applies the Johansen cointegration procedure, as well as the Granger causality test. The results show that Botswana’s inflation dynamics trend positively with all the countries under scrutiny except South Africa in a long-run framework. However, the Granger causality test only proved that Botswana’s inflation lead China’s inflation dynamics. In conclusion, Botswana’s inflation is not driven by other countries’ inflation dynamics.

seekers_:

Abstract—Developing a precise and accurate model of gold price is critical to manage assets because of its unique features. In this paper, artificial neural network (ANN) model have been used for modeling the gold price, and compared with the traditional statistical model of ARIMA (autoregressive integrated moving average). The three performance measures, the coefficient of determination (R2), root mean squared error (RMSE), mean absolute error (MAE), are utilized to evaluate the performances of different models developed. The results show that the ANN model outperforms ARIMA model, in terms of different performance criteria during the training and validation phases.

We discuss risk, preference and valuation in classical economics, which led academics to develop a theory of market prices, resulting in the general equilibrium theories. However, in practice, the decision process does not follow that theory since the qualitative aspect coming from human decision making process is missing. Further, a large number of studies in empirical finance showed that financial assets exhibit trends or cycles, resulting in persistent inefficiencies in the market, that can be exploited. The uneven assimilation of information emphasised the multifractal nature of the capital markets, recognising complexity. New theories to explain financial markets developed, among which is a multitude of interacting agents forming a complex system characterised by a high level of uncertainty. Recently, with the increased availability of data, econophysics emerged as a mix of physical sciences and economics to get the best of both world, in view of analysing more deeply assets' predictability. For instance, data mining and machine learning methodologies provide a range of general techniques for classification, prediction, and optimisation of structured and unstructured data. Using these techniques, one can describe financial markets through degrees of freedom which may be both qualitative and quantitative in nature. In this book we detail how the growing use of quantitative methods changed finance and investment theory. The most significant benefit being the power of automation, enforcing a systematic investment approach and a structured and unified framework. We present in a chronological order the necessary steps to identify trading signals, build quantitative strategies, assess expected returns, measure and score strategies, and allocate portfolios.