Search

This paper describes our work on developing a new business sentiment index using daily newspaper articles. We adopt a recurrent neural network (RNN) with Gated Recurrent Units to predict the business sentiment score of a given text and aggregate the scores to define an index, named S-APIR. An RNN is initially trained on Economy Watchers Survey and then fine-tuned on news texts for domain adaptation. Also, a one-class support vector machine is applied to filter out texts irrelevant to business sentiment. Moreover, we propose a simple yet useful approach to temporally analyzing how much and when any given factor influences the predicted business sentiment. The validity and utility of the proposed approach are empirically demonstrated through a series of experiments on Nikkei Newspaper articles published from 2013 to 2018.

Business sentiment indices play a crucial role in government's policy decisions, production planning in industries, and individual/institutional investment decisions. However, such indices require large-scale surveys, which are costly and laborious to conduct. As an alternative, this paper reports our ongoing work on estimating business sentiment based on abundant, ordinary newspaper articles. Our proposed framework is composed of a recurrent neural network for regression, a one-class support vector machine to filter out topically irrelevant articles, and fine-tuning for domain adaptation. The effectiveness of the framework is empirically demonstrated by experiments on Nikkei newspaper.

Mergers can bring with them not only efficiency gains but also anticompetitive effects. Moreover, when firms announce mergers with limited lead time and information, it is crucial that regulatory bodies be able to judge them appropriately. This paper examines a merger between NKK and Kawasaki Steel that had a deep economic impact. I employ the conventional event study approach using abnormal stock returns. In addition to investigating these two firms and their rival firms, I examine customer firms in four steel-consuming industries in order to paint a broader picture of mergers. I consider the reactions of abnormal returns as measured by twelve patterns, and identify several merger effects. Firms in the steel industry anticipated a significant increase in their future profits as a result of this merger. Of the customer firms, some anticipated an increase in their profits, whereas others expected to suffer losses. However, the statistical significance of their reactions was weak, and, thus, the merger’s effects on customer firms could not be measured accurately. The results suggest that the primary intent of this merger was to improve efficiency. Future studies should investigate the results of the merger in real terms, including its effects on steel prices and steel production, for example, rather than merely in financial terms.