Skip to main content

Table 4 Summary Statistics on Filtered PMG and PML

From: Timing the market: the economic value of price extremes

  Panel A. Monthly Index Data Panel B. Quarterly Index Data
Filtered P MG t P ML t P MG t P ML t
Mean 0.000 0.000 0.000 0.000
Std.dev 1.000 1.000 1.000 1.000
Maxi 3.637 4.915 2.515 3.329
Mini -2.748 -2.343 -2.383 -1.933
Skew -0.006 0.538 0.098 0.740
Kurt 3.212 3.312 2.564 3.363
J-B stat 1.486 41.4 2.505 26.346
Prob 0.476 0.000 0.286 0.000
Auto-Correlation Function (lag)
 ACF(1) 0.003 0.039 0.010 0.008
 ACF(3) 0.039 -0.013 -0.021 -0.057
 ACF(6) -0.026 -0.056 -0.064 -0.038
 ACF(9) -0.008 -0.020 0.017 -0.055
 ACF(12) 0.035 0.053 -0.029 0.047
Q(12) 12.29 12.042 1.092 9.036
Obs 791 791 263 263
  1. Note. To make sure zero sample mean and unit sample standard deviation, the stock returns, good extreme returns and bad extreme returns are first filtered, and then standardized using the following formula
  2. Zt = [Xt ‐ M(Xt)]/S(Xt)
  3. where M(Xt) is the sample mean of Xt, S(Xt) is the sample standard deviation of Xt