- 12/21/18--19:59: Agent-based modelling as a foundation for big data
- 01/04/19--00:58: 結合變數挑選和混頻方法當下預測通膨
title: Agent-based modelling as a foundation for big data abstract: In this article, we propose a process-based definition of big data, as opposed to the size- and technology-based definitions. We argue that big data should be perceived as a continuous, unstructured and unprocessed dynamics of primitives, rather than as points (snapshots) or summaries (aggregates) of an underlying phenomenon. Given this, we show that big data can be generated through agent-based models but not by equationbased models. Though statistical and machine learning tools can be used to analyse big data, they do not constitute a big data-generation mechanism. Furthermore, agentbased models can aid in evaluating the quality (interpreted as information aggregation efficiency) of big data. Based on this, we argue that agent-based modelling can serve as a possible foundation for big data. We substantiate this interpretation through some pioneering studies from the 1980s on swarm intelligence and several prototypical agentbased models developed around the 2000s.
title: Does Institutional Linkage of Bank-MFI Foster Inclusive Financial Development Even in the Presence of MFI Frauds? abstract: Growing reports indicate the presence of frauds in microfinance institutions (MFIs), as it can occur in any organization in countries where there are weak institutions, weak rule of law, and fraudulent behavior of MFI officers for personal gain. While there are increasing calls to launch financial governance of these NGO MFIs, there are concerns as to whether frauds of this nature can damage MFIs' contributions to the credit market, particularly in the bank-linkage program where the NGO MFIs act as third party intermediary. The purpose of this study was to analyze the collusion decisions faced by MFIs and their impact on the bank-linkage program, which has been offered as a solution to help overcome adverse selection and moral hazard problems in the credit market by harnessing local information via MFIs. Our results show that even when there is a chance of collusion between MFI and the borrower, the linkage between MFI and bank can still increase the probability that the borrower puts in full effort, and therefore decreases the probabilities of both credit rationing and strategic default. Such linkage in financing viable projects can make micro-financing more effective in achieving inclusive financial development and thereby poverty reduction in rural areas.
title: MANAGERIAL EFFICIENCY AND TECHNOLOGY GAP RATIO FOR METROPOLITAN AND NON-METROPOLITAN HOTELS IN TAIWAN: APPLICATION OF METAFRONTIER INPUT DISTANCE FUNCTION abstract: The study applies the concept of metafrontier to estimate and compare the managerial efficiencies and technology gap ratios (TGRs) for metropolitan- and non-metropolitan international tourist hotels (ITHs) in Taiwan during the year 1998-2008. The approach that metafrontier model is applied to estimate technical efficiencies for ITHs operating under different technology patterns and the technology gap ratios. Empirical results find that the non-metropolitan hotels have better operating or production efficient scores in managerial side than the metropolitan hotels. Furthermore, the estimations of metafrontier indicate that whether it is the technical efficiency of the common border or the technical gap ratio, hotels in foreign countries, chain cooperative hotels and domestic chain hotels perform better than independent hotels.
title: 結合變數挑選和混頻方法當下預測通膨 authors: 袁瑋成; Yuan, Wei-Cheng
abstract: 本文結合變數挑選與混頻(Mixed frequency)方法，提出兩步驟預測模型，並考慮大量且不同頻率的經濟變數當下預測美國通貨膨脹率。以美國1998年7月到2018年5月的實證結果顯示，加上變數挑選後的混頻模型，其預測表現顯著比無變數挑選的混頻模型好，且僅用少數個挑選出的變數組合預測可以更近一步改善模型的預測表現。而使用不同變數個數組合預測的混頻模型，其預測表現顯著比無混頻模型好，這表示以混頻方法將高頻率變數的資訊納入模型中確實能改善當下預測通膨的預測表現。我們亦發現僅使用少數重要的變數組合預測時，高頻率重要變數對預測表現的影響遠大於低頻率重要變數。此外，考慮不同的穩健性檢驗的結果顯示，本文所提之方法具有穩健性。