Terence Mills' best-selling graduate textbook provides detailed coverage of research techniques and findings relating to the empirical analysis of financial markets. In its previous editions it has become required reading for many graduate courses on the econometrics of financial modelling. This third edition, co-authored with Raphael Markellos, contains a wealth of material reflecting the developments of the last decade. Particular attention is paid to the wide range of nonlinear models that are used to analyse financial data observed at high frequencies and to the long memory characteristics found in financial time series. The central material on unit root processes and the modelling of trends and structural breaks has been substantially expanded into a chapter of its own. There is also an extended discussion of the treatment of volatility, accompanied by a new chapter on nonlinearity and its testing.

7. The Econometric Modelling of Financial Time Series

Effective long short-term memory with differential evolution algorithm for electricity price prediction

Outline: This is the first part of a two-semester sequence course on mathematical finance. In the Fall semester, we will examine the fundamental principles of financial derivatives from both financial and mathematical perspectives, and demonstrate how the mathematical tools from stochastic calculus, differential equations and numerical methods join forces to form an essential part in modern finance. The emphasis of the course is a mathematical understanding of the intrinsic relationships among various financial instruments, which serves as a basis for investment decisions and trading strategies. The central theme of the Fall semester is the classic Black-Scholes-Merton model, and we plan to give a thorough treatment of the original model, with extensive discussions on the practical extensions in response to various disadvantages of the original model. One of the most intuitive and transparent approaches to illustrate that is also extensively used in practice is the binomial tree model. It contains most of the essential idea of the Black-Scholes-Merton methodology, and it can be naturally extended to build more general continuous-time models. Time permitting, we will include as much real life examples as possible to make this a rewarding experience for those who plan to pursue a career in this direction, as well as those who are just intrigued by the subject and its impact on our society.

/pdf/introduction-to-mathematical-finance-1geu6fvnpg.pdf

Introduction to Mathematical Finance

Forecasting stock market indexes is an important issue for market participants, because even a small improvement in forecast accuracy may lead to better trading decisions than those of other participants. Rising interest in deep learning has led to its application in stock market forecasting. However, it is still challenging to use market-size time-series data to predict composite index prices. In this study, we propose a new stock market forecasting framework, NuNet, which can successfully learn high-level features from super-high dimensional time-series data. NuNet is an end-to-end integrated neural network framework consisting of two feature extractor modules, a super-high dimensional market information feature extractor and a target index feature extractor. In addition, we propose a mini-batch sampling technique, trend sampling, which probabilistically samples more recent data when training. Furthermore, we propose a novel regularization method, called column-wise random shuffling, which is a data augmentation technique that can be applied to convolutional neural networks. The experiments are comprehensively carried out in three aspects for three indexes, namely S&P500, KOSPI200, and FTSE100. The results demonstrate that the proposed model outperforms all baseline models. Specifically, for the S&P500, KOSPI200, and FTSE100, the overall mean squared error of our proposed model NuNet(DA, T) is 60.79%, 51.29%, and 43.36% lower than that of the baseline model SingleNet(R), respectively. Moreover, we employ trading simulations with realistic transaction costs. Our proposed model outperforms the buy-and-hold strategy being an average of 2.57 times more profitable in three indexes.

Stock market forecasting with super-high dimensional time-series data using ConvLSTM, trend sampling, and specialized data augmentation

Stock market plays a key role in economical and social organization of a country. Stock market forecasting is highly demanding and most challenging task for investors, professional analyst and researchers in the financial market due to highly noisy, nonparametric, volatile, complex, non-linear, dynamic and chaotic nature of stock price time series. Prediction of stock market is a crucial task and prominent research area in financial domain as investing in stock market involves higher risk. However with the development of computational intelligent methods it is possible to reduce most of the risk. In this survey paper, our focus is on application of computational intelligent approaches such as artificial neural network, fuzzy logic, genetic algorithms and other evolutionary techniques for stock market forecasting. This paper presents an up-to-date survey of existing literature on stock market forecasting based on computational intelligent methods. In this article, the selected papers are organized and discussed according to six main point of view: (1) the stock market analyzed and the related dataset, (2) the type of input variables investigated, (3) the pre-processing techniques used, (4) the feature selection techniques to choose effective variables, (5) the forecasting models to deal with the stock price forecasting problem and (6) performance metrics utilized to evaluate the models. The major contribution of this work is to provide the researcher and financial analyst a systematic approach for development of intelligent methodology to forecast stock market. This paper also presents the outlines of proposed work with the aim to enhance the performance of existing techniques.

Stock Market Forecasting Using Computational Intelligence: A Survey

A new neural network (EBPNN) is developed.An approach to cross-correlations prediction between financial time series.Empirical research is performed in testing the forecasting effect of EBPNN.Forecasting long-term cross-correlations by training short-term cross-correlations.The proposed model is advantageous in increasing the forecasting precision. An improved neural network is developed to predict the cross-correlations between two financial time series. In order to capture the large fluctuations of data set, an exponent back propagation neural network (EBPNN) is introduced in the present work, which information is not only processed locally in each neural unit by computing the dot product between its input vector and its weight vector, but also processed by adding the dot product between its exponential type function of the input vector and its corresponding new weight vector. The proposed prediction model improves the activation function of the neural network, and makes an approach on cross-correlations forecasting with the particular input and output variables. The empirical research is performed in testing the forecasting effect of the EBPNN model and a comparison to back propagation neural network (BPNN). The empirical results show that the EBPNN is advantageous in increasing the predicting precision.

Haiyan Mo

Papers

Exponent back propagation neural network forecasting for financial cross-correlation relationship