The aim of this thesis is to show that news analytics data can be utilised to improve the predictive ability of existing models that have useful roles in a variety of financial applications. The modified models are computationally efficient and perform far better than the existing ones. The new modified models offer a reasonable compromise between increased model complexity and prediction accuracy. I have investigated the impact of news sentiment on volatility of stock returns. The GARCH model is one of the most common models used for predicting asset price volatility from the return time series. In this research, I have considered quantified news sentiment as a second source of information and its impact on the movement of asset prices, which is used together with the asset time series data to predict the volatility of asset price returns. Comprehensive numerical experiments demonstrate that the new proposed volatility models provide superior prediction than the "plain vanilla" GARCH, TGARCH and EGARCH models. This research presents evidence that including news sentiment term as an exogenous variable in the GARCH framework improves the prediction power of the model. The analysis of this study suggested that the use of an exponential decay function is good when the news flow is frequent, whereas the Hill decay function is good only when there are scheduled announcements. The numerical results vindicate some recent findings regarding the utility of news sentiment as a predictor of volatility, and also vindicate the utility of the new models combining the proxies for past news sentiments and the past asset price returns. The empirical analysis suggested that news augmented GARCH models can be very useful in estimating VaR and implementing risk management strategies. Another direction of my research is introducing a new approach to construct a commodity futures pricing model. This study proposed a new method of incorporating macroeconomic news into a predictive model for forecasting prices of crude oil futures contracts. Since these futures contracts are iii iv more liquid than the underlying commodity itself, accurate forecasting of their prices is of great value to multiple categories of market participants. The Kalman filtering framework for forecasting arbitrage-free (futures) prices was utilized, and it is assumed that the volatility of oil (futures) price is influenced by macroeconomic news. The impact of quantified news sentiment on the price volatility is modelled through a parametrized, nonlinear functional map. This approach is motivated by the successful use of a similar model structure in my earlier work, for predicting individual stock volatility using stock-specific news. Numerical experiments with real data illustrate that this new model performs better than the one factor model in terms of accuracy of predictive power as well as goodness of fit to the data. The proposed model structure for incorporating macroeconomic news together with historical (market) data is novel and improves the accuracy of price prediction quite significantly.

This thesis is based on three papers examining Gulf Cooperation Council (GCC) financial markets. The member countries of the GCC are Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates. These countries have transitioned from developing to frontier markets over the past ten years, but there is considerable debate about whether GCC economies are efficient or affected by shocks in oil and other commodity markets. The first paper (chapter 2) considers GCC stock market returns and examines how they are affected by oil price shocks using a bivariate VAR-GARCH(1,1) approach. The conclusion of this essay is that GCC economies are more affected by shocks than are other countries considered for comparison purposes. The second paper (chapter 3) discusses how food prices are affected by oil price shocks, and it examines possible parameter shifts between food and oil that result from four recent events, including renewable fuel policies and the financial crisis. The third paper (chapter 4) uses an empirical approach to compare a least squares model and a non-linear Markov switching model to measure the effect of newspaper sentiment on stock market performance. The results indicate that all information is important to stock market investors and that non-linear models are better predictors of stock market performance then linear models when using data from newspaper articles. Chapter 5 offers some final conclusions and remarks.

Deep penetration of personal computers, data communication networks, and the Internet has created a massive platform for data collection, dissemination, storage, and retrieval. Large amounts of textual data are now available at a very low cost. Valuable information, such as consumer preferences, new product developments, trends, and opportunities, can be found in this large collection of textual data. Growing worldwide competition, new technology development, and the Internet contribute to an increasingly turbulent business environment. Conducting surveillance on this growing collection of textual data could help a business avoid surprises, identify threats and opportunities, and gain competitive advantages.Current text mining approaches, nonetheless, provide limited support for conducting surveillance using textual data. In this dissertation, I develop novel text quantification approaches to identify useful information in textual data, effective anomaly detection approaches to monitor time series data aggregated based on the text quantification approaches, and empirical evaluation approaches that verify the effectiveness of text mining approaches using external numerical data sources.In Chapter 2, I present free-text chief complaint classification studies that aim to classify incoming emergency department free-text chief complaints into syndromic categories, a higher level of representation that facilitates syndromic surveillance. Chapter 3 presents a novel detection algorithm based on Markov switching with jumps models. This surveillance model aims at detecting different types of disease outbreaks based on the time series generated from the chief complaint classification system.In Chapters 4 and 5, I studied the surveillance issue under the context of business decision making. Chapter 4 presents a novel text-based risk recognition design framework that can be used to monitor the changing business environment. Chapter 5 presents an empirical evaluation study that looks at the interaction between news sentiment and numerical accounting earnings information. Chapter 6 concludes this dissertation by highlighting major research contributions and the relevance to MIS research.

本篇研究專注於首次公開發行公司上市櫃初始交易日之異常報酬與新聞情緒兩 者間之關係。本研究建立情緒字典以判別新聞之正負情緒,並過濾出與首次公開發 行有關之新聞,利用本研究建立之情緒字典以過濾出正負情緒之詞組。利用正負情 緒詞組數量計算出三種新聞情緒變數,並採實證研究方法檢測三種新聞情緒變數與 首次公開發行公司之初始交易日之異常報酬兩者間之關係。根據本研究之實證結果, 發現初始交易日之前的新聞能影響首次公開發行之異常報酬,而相關新聞之情緒語 調亦和異常報酬有關。此外,本研究亦檢測三種情緒變數和三種傳統變數之交乘項 對異常報酬之影響,發現公司規模大小與首日交易量與情緒變數之交乘項會對初始 交易日之異常報酬有影響。總言論之,本研究對新聞會影響首次公開發行初始交易 日之異常報酬提供了實證證據。 / This study focuses on the relation between IPOs’ abnormal returns on initial trading days and news sentiment. To identify the tone of news, sentiment dictionary was established for this study, and news regarding IPO firms was picked out to count positive and negative words and phrases based on the sentiment dictionary. Using quantities of positive and negative words and phrases, three news variables were adopted and calculated. And linear regression was utilized to investigate the relation between IPOs’ abnormal returns on initial trading days and news sentiment. According to empirical results, I find that news prior to the IPO’s initial trading day can affect IPOs’ abnormal returns. The number of negative words and phrases is negatively related to the abnormal returns; the tone of news is positively related to the abnormal returns. Furthermore, I also investigated whether interaction terms of news variables and three control variables are related to abnormal returns on IPOs’ initial trading days. I find that interaction terms of the natural logarithm of firm size and two news variables and interaction terms of the natural logarithm of first-day trading volume and two news variables are related to abnormal returns. Overall, there is evidence that news can influence IPOs’ abnormal returns on initial trading days.

This thesis explores to what extent Multinomial Naive Bayes (MNB) and Support Vector Machines (SVM) classifiers can be used to determine the polarity of news, specifically the news coverage of Sweden by the Russian state-funded news outlets RT and Sputnik. Three experiments are conducted.  In the first experiment, an MNB and an SVM classifier are trained with the Large Movie Review Dataset (Maas et al., 2011) with a varying number of samples to determine how training data size affects classifier performance.  In the second experiment, the classifiers are trained with 300 positive, negative, and neutral news articles (Agarwal et al., 2019) and tested on 95 RT and Sputnik news articles about Sweden (Bengtsson, 2019) to determine if the domain specificity of the training data outweighs its limited size.  In the third experiment, the movie-trained classifiers are put up against the domain-specific classifiers to determine if well-trained classifiers from another domain perform better than relatively untrained, domain-specific classifiers.  Four different types of feature sets (unigrams, unigrams without stop words removal, bigrams, trigrams) were used in the experiments. Some of the model parameters (TF-IDF vs. feature count and SVM’s C parameter) were optimized with 10-fold cross-validation.  Other than the superior performance of SVM, the results highlight the need for comprehensive and domain-specific training data when conducting machine learning tasks, as well as the benefits of feature engineering, and to a limited extent, the removal of stop words. Interestingly, the classifiers performed the best on the negative news articles, which made up most of the test set (and possibly of Russian news coverage of Sweden in general).