Session Chair: Krisda Nimmanunta, NIDA Business School

Paper Presentations:

Financial Co-integration of LDCs of Asia and the Pacific with China (Paper withdrawn / presentation cancelled)
Vipul Kumar Singh, National Institute of Industrial Engineering

This paper explores the level of financial integration among the LDCs of the Asia and the Pacific region, and also ascertains their level of financial integration with China. For same, we examined the existence of long-term integration between the exchange rates of the LDCs by creating data sets namely LDCs of SAARC, LDCs of ASEAN, LDCs of Pacific, LDCs of SAARC & ASEAN, LDCs of ASEAN & Pacific, LDCs of SAARC & Pacific, and LDCs of Asia and the Pacific. In addition, we have also tested the cointegration of these 7 groups with China. The analysis has been carried out using the following models: Johansen multivariate co-integration with error correction model, advanced Gregory-Hansen cointegration test, and multivariate DCC GARCH model.

Effect of Commission Payment on Rating Inflation
Kittiphod Charoontham, NIDA Business School
Thunyarat Amornpetchkul, NIDA Business School

This study investigates whether the commission scheme can be used to discourage CRA’s rating inflation under an issuer-pay model. We model a setting where an issuer can request a level of information accuracy; the CRA can decide rating reported regime. The CRA is offered an upfront to cover information cost as well as a commission tied to the project outcome. We find that commission, with the issuer’s requested level of information accuracy, can induce the CRA to give truthful ratings in good and bad. In good market, the CRA reports truthfully if the commission is relatively small, and inflates ratings otherwise. In bad market, the CRA always reports truthfully. Our results illustrate how the payment scheme and the issuer’s actions can affect the CRA’s behavior.

Stock Price Analysis under Extreme Value Theory
Paul Inpong Louangrath, Bangkok University

The objective of this paper is to provide a practical tool for stock price evaluation and forecasting under Extreme Value Theory (EVT). Three existing models are reviewed: Mordern Portfolio Theory, Black-Scholes, and Jarrow-Rudd models. These models are not effective tools where option contract is not available. The data consist of the daily close price of 30 days from 100 companies in the SET100 index. It was found that Thailand’s SET100 consists of two groups of stocks according to price distribution, namely Weibull and Fréchet. This finding rejects the assertion that most financial data are fat-tailed. It implies that investors face two categories of stocks: low and high price volatility. Sector diversification becomes secondary. Price volatility index is a better indicator for risk management.

On a Multi-stage Estimation of Copula-based Regime-switching Models for Financial Time Series
Kridsda Nimmanunta, NIDA Business School

This paper addresses the issue of using multi-stage likelihood estimation in Markov regime-switching models, which is seemingly impossible due to the (additive) inseparability of the log-likelihood function. Using a modi_ed EM algorithm proposed by Arcidiacono and Jones (2003), this paper shows an e_ective way to estimate copula-based Markov regime-switching models. On the convergence, the algorithm provides an estimator analogous to that of the two-step approach, where the di_erence is that the latter is the special case of the multi-stage EM estimator when there is only one regime. This paper also proposes a copula-based Markov regime-switching GARCH model where both the copula and margins are regime-switching. The salient feature of this model is that the unobservable regimes can be either co-switching or counter-switching with each other.