Papers by Krisada Khruachalee
Journal of Humanities and Social Sciences, Dec 28, 2020
The objective of this study is to classify the company listed in the Stock Exchange of Thailand b... more The objective of this study is to classify the company listed in the Stock Exchange of Thailand by applying the Principle Component Analysis (PCA) to overcome the multicollinearity problem in the purpose of extracting the common factors used in the Logistic regression model. The data set are gathered from the companies’ annual financial statements registered in the Stock Exchange of Thailand since 2002 to 2016. The data set consists of 101 companies which there are only 29 distressed companies that have been defaulted on corporate bonds and there are 72 non-distressed companies that have not been defaulted on corporate bonds. The logodds score are calculated based on the predicted default probability generated from the Logistic regression whose variables are integrated with the 4 statistically significant common factors. These scores are used to classify the company’s risk level. The overall accuracy of the predicted model is 92.3% which we can group the companies’ default probabili...
The objectives of this study are to forecast and model the Asian foreign exchange rate by using t... more The objectives of this study are to forecast and model the Asian foreign exchange rate by using the extension of time series analysis technique proposed by BoxJenkins (1970), which it is universally known as an Autoregressive Integrated Moving Average with Explanatory Variable or “ARIMAX”. This empirical study selectively gathers the foreign exchange rate of the Asian countries since September 2015 to March 2017. The sample consists of 4 major Asian currencies that are actively traded in the foreign exchange market including Japanese Yen (JPY), Chinese Yuan (CNH), Singapore Dollar (SGD), and Malaysia Ringgit (MYR). These currencies are specifically denominated in Thai Baht (THB). In order to overcome a misspecification problem, Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) are used as a criterion to select the forecasting model. The forecasting performance of each model has been competed together with a classical ARIMA and a random walk model. The finding shows t...
Lobachevskii Journal of Mathematics
In this article we introduce the new, two-parameter partial-geometric distribution (PG) that cont... more In this article we introduce the new, two-parameter partial-geometric distribution (PG) that contains both geometric and first success distributions as a particular case. Some probability and statistical properties of the proposed distribution are discussed, including probability mass function, mean, variance, moment generating function, and probability generating function. We propose the method of maximum likelihood for estimating the model's parameters, and apply the PG distribution to two real datasets to illustrate the flexibility of the proposed distribution. We found the PG distribution is more dynamic than the geometric distribution in the sense that it can be applied to the under-dispersed data. The PG distribution also performs well with a goodness of fit test and some other model selection characteristics for model fitting of these two datasets. Thus, the PG distribution can be applied as an alternative model for the analysis of discrete data.
ABAC Journal, Jul 30, 2021
The uncertainty of return on investment is a major concern for the vast majority of investors. Un... more The uncertainty of return on investment is a major concern for the vast majority of investors. Under normal market conditions, a portfolio's risk exposure over a specific time frame with a predetermined confidence level can be measured. Since a portfolio's return is rarely characterized by the assumption of a multivariate normal distribution, the use of normality Value-at-Risk (VaR) plays a crucial role in risk mitigation, but can generate an undesirable measure of risk exposure for portfolio investment. With a dynamic tool in modeling multivariate distribution regardless of the assumption of joint normality, applying a copula is a practical alternative choice for extracting a cumulative joint distribution for a portfolio's return. The applications in this work are illustrated by the portfolios of the four largest and the four smallest market capitalization stocks in the tourism and hospitality sector. It was found that the portfolio returns of the large and small market capitalization portfolios were characterized by logistic and Student's t distributions respectively. Consequently, the VaR and conditional VaR based on the Gaussian copula, could be used to characterize and estimated the distributions of the respective portfolio returns according to the logistic and Student's t distributions. The conditional VaR of the large and small market capitalization portfolios calculated from the copula method provides a slightly higher level of risk than the conditional VaR and the VaR with the assumption of a multivariate normal distribution. Moreover, the small market capitalization portfolio provides slightly higher values of VaR and CVaR than the large market capitalization portfolio for all assumptions of VaR. Therefore, the use of conditional VaR 1 ,* Mr. Krisada Khruachalee obtains two master's degrees-Master of Science
Uploads
Papers by Krisada Khruachalee