Table 7

Removing effect of investor sentiment: Monthly excess stock market returns against conditional variances and implied variance with uncertainty

Modela1b1a2b2Adj. R2 (%)
Panel A: Rolling window model
(1)-E0.0020 (0.7564)3.8492*** (4.5014)0.0089* (1.7045)−5.7503*** (−7.7159)8.640
(1)-V−7.12 × 10–4 (−0.2854)4.3781*** (6.3711)0.0142*** (3.7523)−6.4427*** (−11.3850)10.498
Panel B: MIDAS
(2)-E1.59 × 10–4 (0.0555)3.2272*** (5.9357)0.0107** (2.1665)−4.9393*** (−8.3578)12.596
(2)-V−0.0015 (−0.5874)3.1987*** (5.3412)0.0146*** (3.9200)−4.9571*** (−8.2909)12.819
Panel C: GARCH(1,1)
(3)-E4.81 × 10–4 (0.1732)3.7434*** (6.2314)0.0091* (1.8885)−5.3781*** (−8.5842)7.239
(3)-V−0.0018 (−0.6836)3.9706*** (6.0568)0.0137*** (3.6034)−5.7159*** (−8.8836)8.097
Panel D: GJR-GARCH(1,1)
(4)-E−0.0019 (−0.6117)5.4890*** (6.1301)0.0137*** (2.8953)−7.7743*** (−7.5828)7.938
(4)-V−0.0044 (−1.4859)5.7892*** (5.9516)0.0189*** (4.4732)−8.2731*** (−8.3799)9.255
Panel E: Implied variance
(5)-E−0.0042 (−1.1450)0.4054*** (6.2509)0.0138** (2.0619)−0.5294*** (−3.6461)4.232
(5)-V−0.0064* (−1.8488)0.4188*** (6.2855)0.0191*** (3.4688)−0.5658*** (−4.1600)4.991

Note(s): The table reports estimates from the regressions of monthly excess stock market returns against conditional variances, and implied variance. Models (1)-E to (5)-E ((1)-V to (5)-V) are estimates from the regressions with a dummy variable of FEPU (FVOV) after removing the effect of investor sentiment. Newey and West (1987) corrected t-statistics are reported in parentheses. *, ** and *** represent significance at the 10, 5 and 1% levels, respectively

or Create an Account

Close subscription notice
Close access options