Transformations Foreign Direct Investment Data in Thailand: Case Study of ASEAN Countries

Abstract

Foreign direct investment (FDI) data in Thailand of the countries in Asian (12 months) are symmetry, left-skewed or right-skewed. The normally distributed data are an important assumption in the statistical analysis methods. If the data do not correspond with the assumption, data transformation will be used. The objective of this research is to compare six transformation methods: square root transformation, logarithm base ten transformation, inverse transformation, Box-Cox transformation, Exponential transformation, and Yeo-Johnson transformation. The results reveal that square root transformation can transform FDI data of Laos and Myanmar to symmetry and normally distributed data. Logarithm base ten transformation can transform FDI data of Cambodia and Myanmar to symmetry and normally distributed data. Inverse transformation can transform FDI data of Vietnam to near symmetry and normally distributed data. Box-Cox transformation can transform FDI data of Cambodia Laos Myanmar Vietnam and Singapore to symmetry and normally distributed data. Exponential transformation can transform FDI data of Cambodia Laos and Myanmar to symmetry and normally distributed data. Yeo-Johnson transformation can transform FDI data of Cambodia Laos Myanmar and Singapore to symmetry and normally distributed data. Reflect technique can adjust left-skewed data to right-skewed data. If data are both negative and positive value, they should be adjusted to the positive value before transforming data.