Tiffany Hui-Kuang Yu

Ratio-based lengths of intervals to improve fuzzy time series forecasting

The objective of this study is to explore ways of determining the useful lengths of intervals in fuzzy time series. It is suggested that ratios, instead of equal lengths of intervals, can more properly represent the intervals among observations. Ratio-based lengths of intervals are, therefore, proposed to improve fuzzy time series forecasting. Algebraic growth data, such as enrollments and the stock index, and exponential growth data, such as inventory demand, are chosen as the forecasting targets, before forecasting based on the various lengths of intervals is performed. Furthermore, sensitivity analyses are also carried out for various percentiles. The ratio-based lengths of intervals are found to outperform the effective lengths of intervals, as well as the arbitrary ones in regard to the different statistical measures. The empirical analysis suggests that the ratio-based lengths of intervals can also be used to improve fuzzy time series forecasting.

A bivariate fuzzy time series model to forecast the TAIEX

Fuzzy time series models have been applied to forecast various domain problems and have been shown to forecast better than other models. Neural networks have been very popular in modeling nonlinear data. In addition, the bivariate models are believed to outperform the univariate models. Hence, this study intends to apply neural networks to fuzzy time series forecasting and to propose bivariate models in order to improve forecasting. The stock index and its corresponding index futures are taken as the inputs to forecast the stock index for the next day. Both in-sample estimation and out-of-sample forecasting are conducted. The proposed models are then compared with univariate models as well as other bivariate models. The empirical results show that one of the proposed models outperforms the many other models.

A Multivariate Heuristic Model for Fuzzy Time-Series Forecasting

Fuzzy time-series models have been widely applied due to their ability to handle nonlinear data directly and because no rigid assumptions for the data are needed. In addition, many such models have been shown to provide better forecasting results than their conventional counterparts. However, since most of these models require complicated matrix computations, this paper proposes the adoption of a multivariate heuristic function that can be integrated with univariate fuzzy time-series models into multivariate models. Such a multivariate heuristic function can easily be extended and integrated with various univariate models. Furthermore, the integrated model can handle multiple variables to improve forecasting results and, at the same time, avoid complicated computations due to the inclusion of multiple variables.

A neural network-based fuzzy time series model to improve forecasting

Neural networks have been popular due to their capabilities in handling nonlinear relationships. Hence, this study intends to apply neural networks to implement a new fuzzy time series model to improve forecasting. Differing from previous studies, this study includes the various degrees of membership in establishing fuzzy relationships, which assist in capturing the relationships more properly. These fuzzy relationships are then used to forecast the stock index in Taiwan. With more information, the forecasting is expected to improve, too. In addition, due to the greater amount of information covered, the proposed model can be used to forecast directly regardless of whether out-of-sample observations appear in the in-sample observations. This study performs out-of-sample forecasting and the results are compared with those of previous studies to demonstrate the performance of the proposed model.

Entrepreneurship, process innovation and value creation by a non‐profit SME

Purpose – By using three key factors – namely, funding, stakeholders, and legitimacy – this study seeks to analyse the successful entrepreneurial experiences of a non‐profit small to medium‐sized enterprise: the Taiwan EBook Supply Cooperative Limited (TEBSCo).Design/methodology/approach – The paper takes the form of a case study.Findings – From a legitimacy perspective, TEBSCo is the only registered organisation facilitating e‐book consortia in Taiwan. From a stakeholder perspective, TEBSCo is managed by a board of directors, who are elected from the member representatives. In addition to creating value for its members, TEBSCo also creates value for non‐members and vendors. Its major funding is from annual membership fees. TEBSCos innovation process, as a collective entrepreneurial activity in a non‐profit SME, creates intangible as well as tangible value. The successful experiences of TEBSCo can be used as examples for new entrants.Originality/value – TEBSCo is the only registered organisation facilita...

An Advanced Approach to Forecasting Tourism Demand in Taiwan

Abstract Forecasting has been considered important in a service industry. Many techniques have been applied to improve forecasting results. This study intends to apply a neural network based fuzzy time series model to forecast the international tourist numbers arriving in Taiwan. Neural network is good at handling nonlinear data. On the other hand, the fuzzy time series models have been applied to time series problems in various domains and have been shown to outperform some conventional models. The tourist numbers arriving in Taiwan show a nonlinear characteristic with a structural break. And the application of the neural network based fuzzy time series model is expected to outperform some other models in forecasting these tourist numbers.

Entrepreneurial firms' wealth creation via forecasting

Wealth creation is critical to the performance of entrepreneurial firms. The two major issues that entrepreneurial firms face are as to when to issue the initial public offering (IPO) and how to invest in the stock market. Stock market forecasting can facilitate the provision of financial services for entrepreneurial firms in relation to both issues. Hence, this study proposes a novel neural network multivariate model to forecast stock markets. The proposed model can assist in deciding the timing of an IPO for the entrepreneurial firms and when to invest in the stock market.

An innovative regime switching model to forecast Taiwan tourism demand

The tourism industry has become a major part of economic development for many countries. These countries have greatly invested in tourism to attract more tourist arrivals. Hence, the need for more accurate forecasts of tourism demand is important. Various approaches have been applied to forecast tourism demand of different countries. However, tourism demands tend to be imprecise and their trends nonlinear. In addition, there may be drastic changes in the tourism demand time series. To properly handle these problems, this study proposes an innovative forecasting model to detect the regime switching properly and to apply fuzzy time-series model to forecast. The monthly tourist arrivals to Taiwan will be used as forecasting target. The analysis by the proposed model will be validated by the major events as well as previous studies.

Internet software and services: past and future

The Internet has been extremely popular because of its many unique and powerful characteristics, such as ubiquity and global reach. Hence, it is interesting to study whether Internet companies can be sustained through a bubble and even after a bubble. This study therefore compares the Internet companies in the USA with other high-tech companies listed on the National Association of Securities Dealers Automated Quotations (NASDAQ) over a period covering the years from 1995 to 2006. First, we check whether there has been a bubble for these Internet companies, and we then try to compare the pre- and post-bubble periods to ascertain whether any of their stocks have been ‘overvalued’. Conclusions are drawn regarding the respective futures of these Internet companies.

Forecasting tourism demand by fuzzy time series models

Purpose – This study aims to adapt a neural network based fuzzy time series model to improve Taiwans tourism demand forecasting.Design/methodology/approach – Fuzzy sets are for modeling imprecise data and neural networks are for establishing non‐linear relationships among fuzzy sets. A neural network based fuzzy time series model is adapted as the forecasting model. Both in‐sample estimation and out‐of‐sample forecasting are performed.Findings – This study outperforms previous studies undertaken during the SARS events of 2002‐2003.Research limitations/implications – The forecasting model only takes the observation of one previous time period into consideration. Subsequent studies can extend the model to consider previous time periods by establishing fuzzy relationships.Originality/value – Non‐linear data is complicated to forecast, and it is even more difficult to forecast nonlinear data with shocks. The forecasting model in this study outperforms other studies in forecasting the nonlinear tourism demand...