Table 2

Sensitivity analysis of entropy-resilience estimates across rolling windows and histogram bins

Rolling window (weeks)Histogram binsMean entropyStd entropyMean resilienceStd resilienceMin resilienceMax resilienceBest lead-lag (weeks)Best lead-lag correlation
850.8080.1630.1920.1630.0310.7669−0.145
880.7360.1280.2640.1280.0830.81911−0.179
8100.7090.1240.2910.1240.0970.8369−0.152
8120.6690.1120.3310.1120.1630.8489−0.190
1250.8060.2000.1940.2000.0130.8222−0.111
1280.7760.1550.2240.1550.0270.8623−0.103
12100.7570.1520.2430.1520.0210.8759−0.114
12120.7350.1430.2650.1430.0930.8857−0.140
1650.7820.2440.2180.2440.0050.8552−0.114
1680.7730.1940.2270.1940.0160.88810−0.125
16100.7640.1860.2360.1860.0220.89811−0.143
16120.7530.1770.2470.1770.0240.90610−0.155
2050.7440.2740.2560.2740.0080.87710−0.127
2080.7560.2160.2440.2160.0220.9052−0.120
20100.7530.2120.2470.2120.0230.91410−0.145
20120.7490.2080.2510.2080.0180.9209−0.146

Note(s): The table reports sensitivity results for alternative rolling-window lengths L and histogram-bin counts n. For each specification, normalized Shannon entropy Ht(N), resilience Rt=1Ht(N), and the strongest negative lead-lag correlation between Rt and rt+k are computed. The lead-lag horizon is measured in weeks

or Create an Account

Close Modal
Close Modal