臺大管理論叢 NTU Management Review VOL.29 NO.1

Earnings Informativeness of Long-Lived Assets Impairment Recognized and Reversals 222 Table 4 Pearson/Spearman Correlation Matrix for Related Variables R t X t-1 X t X t3 R t3 IM t SIZE t LEV t MB t R t -0.065 a 0.338 a 0.269 a -0.244 a -0.049 a 0.008 0 -0.050 a 0.398 a X t-1 0.081 a 0.512 a 0.385 a 0.070 a -0.127 a 0.154 a -0.160 a 0.118 a X t 0.416 a 0.638 a 0.599 a 0.035 b -0.187 a 0.149 a -0.109 a 0.233 a X t3 0.274 a 0.560 a 0.657 a 0.407 a -0.076 a 0.082 a 0.013 0.141 a R t3 -0.240 a 0.160 a 0.119 a 0.470 a -0.014 -0.041 a 0.007 -0.211 a IM -0.069 a -0.129 a -0.165 a -0.089 a -0.034 -0.077 a 0.063 a -0.026 SIZE 0.059 a 0.180 a 0.200 a 0.153 a 0.011 -0.008 a 0.121 a 0.006 LEV -0.058 a -0.094 a -0.088 a -0.023 -0.020 0.089 a 0.136 a -0.064 a MB 0.466 a 0.197 a 0.332 a 0.194 a -0.275 a -0.096 a 0.026 -0.076 a Legends: 1. R t : a firm’s ex-dividend annual stock return in year t . X t-1 : the earnings per share in year t -1, deflated by the stock price at the beginning of year t . X t : the earnings per share in year t , deflated by the stock price at the beginning of year t . X t3 : the sum of earnings per share for year t +1 through t +3, deflated by the stock price at the beginning of year t . R t3 : the annually compounded returns for year t +1 through t +3. IM t : impairment loss in year t , deflated by the total assets at the beginning of year t . LEV t : firm’s leverage in year t . MB t : market-to-book ratio in year t . SIZE t : natural logarithm of book value of total assets in year t . 2. “a” and “b” denote the significance on the 1% and 5% levels respectively, based on two-tailed tests. 3. The upper triangular of matrix presents Pearson correlation coefficients, and the lower triangular of matrix presents Spearman correlation coefficients. 4.2 Regression Results The estimation process of this study begins with the least-squares regression of the pooled data followed by an assessment of the validity of the pooled model’s assumption of a single, overall intercept term. The Lagrange Multiplier Statistic (LM test) rejects the pooled model (LM = 22.92 > χ (5) in Reg. (1) and LM = 25.29 > χ (10) in Reg. (2), which implies heterogeneous intercepts), thus the panel data model offers a more powerful approach. Subsequently, the estimation proceeds to the panel data analysis and a choice between the fixed effect and a random effect. The Hausman specification test (Hausman, 1978) reveals the potential for omitted variable bias and the importance of firm-specific effects in this setting ( χ 2 = 1,271.80 in Reg. (1) and χ 2 = 1,274.27 in Reg. (2)). We anticipate the need to use the fixed-effect unbalanced-panel approach (Greene, 2004) to examine the influence of asset impairment recognition on earnings informativeness.