本セミナーの企画・運営は、浅川和宏先生(経営管理研究科)です。
お問合せは、浅川先生までお願いいたします。
発表者 | Professor Yi Li (University of Sydney) |
---|---|
日時 | 2023年10月31日(火)18:00~19:00 |
テーマ | Exploring the Impact of Perceived Environmental Uncertainty Diversification on Emerging Market Multinational Enterprises’ Foreign Divestments: A Research with Natural Language Processed Topic Modelling |
要旨 | Abstract: This study examines the impact of perceived environmental uncertainty (PEU) on the subsidiary-level divestments of multinational enterprises (MNEs), under the context of deglobalization. The PEU perspective focuses on how members within an organization perceive and interpret the uncertainty in their external environment, since PEU is a function of both the perceiver and the environment. While existing literature mainly focus on the managerial level of PEU, it largely ignores the diverse sources of PEU at the organization level. Accordingly, we develop a PEU theme framework and capture it by adopting a novel machine learning method-dynamic topic modelling (DTM) to extract latent themes with time evolution. By employing this natural language processing (NLP) technique, the voluminous size of textual data can be analyzed in a short period. As an updated version of the Latent Dirichlet allocation model, the mechanism behinds the DTM is analyzing the probability distribution of both the keywords under the possible topics and the topics within all documents in the textual data set through summarizing the co-occurrence and frequency of words. Building on the PEU themes framework, we argue that when PEU are more concentrated on few sources, the focal MNE is more likely to conduct deglobalization actions - foreign subsidiary divestment, as compared to perceiving diverse environmental uncertainty. This deglobalization effect tends to be radical rather than incremental, and MNEs with larger size experience greater deglobalization effect at a subsidiary level. We extend the PEU perspective literature by bringing in the concept of PEU diversification, with conceptual advancement from a managerial level to an organizational level. Our panel data for regressions include 26,975 firm-year observations over eleven years (2005-2016), and 7,628 observations in Heckman stage 1 after excluding missing data of variables drawn from 402 publicly listed Chinese firms. |
場所 | 対面:エグゼクティブセミナールーム(協生館5階) Zoom ※対面とオンラインのハイブリッド形式で実施します。 ※事前申込等はありませんので、対面参加を希望される方は、直接、協生館4階の階段教室1にお越しください。 また、オンラインでの参加を希望される方は、時間になりましたら、直接Zoomに入室してください。 |