An error correction factor model of term structure slopes in international swap markets

Commonality in trading activity and futures‐cash basis: Evidence from the Taiwan futures and stock markets

“…Totally, the first three principal components explain approximately 92% of the total variation. To further identify the first principal component, this study uses linear projections of each trading activity on the first principal component (see Abad & Novales, 2005). Specifically, for each trading activity of each investor type, the difference in trading volume regresses on the first principal component.…”

“…As indicated by Abad and Novales (), principal components are defined up to a scale factor, so that the same linear combinations, multiplied by a given real number, or changed in sign, would contain the same information. Thus, in order to compare the VAR results regarding the impact of all types of investors' trading activities on the futures‐cash basis, if the regression coefficient of the identified common factor is negative, the −1 } common factor is used to study the effect of the common factor on the futures‐cash basis in VAR analysis.…”

“…Finally, the results of the impulse response functions show that the first principal component of trading activity as well as mutual funds' futures trading have a greater impact on the futures-cash basis than other common factors and other investor types. 1 As indicated by Abad and Novales (2005), principal components are defined up to a scale factor, so that the same linear combinations, multiplied by a given real number, or changed in sign, would contain the same information. Thus, in order to compare the VAR results regarding the impact of all types of investors' trading activities on the futures-cash basis, if the regression coefficient of the identified common factor is negative, the Ϫ1 ϫ common factor is used to study the effect of the common factor on the futures-cash basis in VAR analysis.…”