1. Sample description
The empirical analyses are based on Eq. (8); each observation is a firm-month. The construction of the sample begins
with the monthly return residuals for all stocks from January 1983 to December 2008.12 Each return residual is then
matched with the monthly market factors, MKT, SMB, HML, and LIQ, for the same month. The determinants of liquidity risk
and market risk then are matched to each firm-month with a time lag of at least one month to reduce the likelihood of
reverse causality. In particular, for all variables that use financial statement data from the Compustat Annual database
(e.g., Earnings precision, Accruals quality, and Book-to-market), their values at the most recent prior fiscal year-end are
matched to the firm-month with a lag of four months; the lag is to ensure that the financial statement data is publicly
available. For example, a financial statement variable measured as of 31 December 2000 is matched to each firm-month
from May 2001 to April 2002. All the other variables (e.g., Analyst consensus, Liquidity, and Volume) are available monthly
and are hence lagged by one month.
To maintain a constant sample in th
1. Sample descriptionThe empirical analyses are based on Eq. (8); each observation is a firm-month. The construction of the sample beginswith the monthly return residuals for all stocks from January 1983 to December 2008.12 Each return residual is thenmatched with the monthly market factors, MKT, SMB, HML, and LIQ, for the same month. The determinants of liquidity riskand market risk then are matched to each firm-month with a time lag of at least one month to reduce the likelihood ofreverse causality. In particular, for all variables that use financial statement data from the Compustat Annual database(e.g., Earnings precision, Accruals quality, and Book-to-market), their values at the most recent prior fiscal year-end arematched to the firm-month with a lag of four months; the lag is to ensure that the financial statement data is publiclyavailable. For example, a financial statement variable measured as of 31 December 2000 is matched to each firm-monthfrom May 2001 to April 2002. All the other variables (e.g., Analyst consensus, Liquidity, and Volume) are available monthlyand are hence lagged by one month.To maintain a constant sample in th
การแปล กรุณารอสักครู่..