Day-of-the-week returns and mood: an exterior template approach

Zilca, Shlomo

doi:10.1186/s40854-017-0079-4

Financial Innovation

Table 3 Day-of-the-week effect and mood

From: Day-of-the-week returns and mood: an exterior template approach

	VW	EW	1	2	3	4	5	6	7	8	9	10
Panel A: 2007 scores
λ₀	-0.2899	-0.6905	-0.9099	-0.8949	-0.8544	-0.8172	-0.7357	-0.6647	-0.5852	-0.5039	-0.4621	-0.2288
Newey-West S.E.	0.0502	0.0417	0.0446	0.0412	0.0424	0.0439	0.0461	0.0484	0.0496	0.0489	0.0494	0.0520
t-statistic	-5.77	-16.56	-20.40	-21.72	-20.15	-18.62	-15.96	-13.73	-11.80	-10.30	-9.35	-4.40
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
λ₁	0.0467	0.1114	0.1467	0.1443	0.1378	0.1318	0.1186	0.1072	0.0944	0.0813	0.0745	0.0369
Newey-West S.E.	0.0079	0.0063	0.0067	0.0061	0.0063	0.0066	0.0070	0.0073	0.0075	0.0075	0.0076	0.0082
t-statistic	5.94	17.80	22.03	23.69	21.84	20.09	17.04	14.60	12.52	10.87	9.83	4.51
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
R²	0.0023	0.0189	0.0269	0.0311	0.0274	0.0243	0.0179	0.0137	0.0101	0.0075	0.0063	0.0013
Adjusted R²	0.0022	0.0188	0.0268	0.0310	0.0274	0.0243	0.0178	0.0136	0.0100	0.0075	0.0063	0.0013
Proportion of the effect explained by mood	39.5%	73.8%	87.4%	84.5%	81.9%	79.8%	72.4%	67.3%	62.3%	55.2%	53.4%	30.8%
Panel B: Farber scores
λ₀	-0.3571	-0.7526	-0.9250	-0.9355	-0.9060	-0.8740	-0.8047	-0.7345	-0.6557	-0.5772	-0.5319	-0.2963
Newey-West S.E.	0.0521	0.0428	0.0460	0.0423	0.0436	0.0451	0.0474	0.0496	0.0508	0.0502	0.0506	0.0540
t-statistic	-6.85	-17.58	-20.11	-22.12	-20.78	-19.38	-16.98	-14.81	-12.91	-11.50	-10.51	-5.49
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
λ₁	0.0576	0.1213	0.1491	0.1508	0.1460	0.1409	0.1297	0.1184	0.1057	0.0930	0.0857	0.0478
Newey-West S.E.	0.0081	0.0064	0.0068	0.0062	0.0064	0.0067	0.0071	0.0074	0.0077	0.0076	0.0077	0.0084
t-statistic	7.12	19.10	21.83	24.28	22.78	21.16	18.35	15.91	13.82	12.24	11.14	5.66
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
R²	0.0033	0.0212	0.0263	0.0321	0.0292	0.0263	0.0202	0.0158	0.0120	0.0093	0.0079	0.0021
Adjusted R²	0.0032	0.0212	0.0262	0.0321	0.0291	0.0263	0.0201	0.0158	0.0119	0.0093	0.0079	0.0020
Proportion of the effect explained by mood	56.7%	82.9%	85.3%	87.3%	87.0%	86.3%	81.9%	77.7%	73.9%	68.4%	66.9%	48.8%
Panel C: Simple average
λ₀	-0.3266	-0.7297	-0.9287	-0.9260	-0.8904	-0.8553	-0.7788	-0.7073	-0.6272	-0.5462	-0.5022	-0.2649
Newey-West S.E.	0.0513	0.0425	0.0455	0.0419	0.0432	0.0447	0.0469	0.0493	0.0505	0.0498	0.0503	0.0532
t-statistic	-6.37	-17.17	-20.41	-22.10	-20.61	-19.13	-16.61	-14.35	-12.42	-10.97	-9.98	-4.98
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
λ₁	0.0526	0.1176	0.1497	0.1493	0.1436	0.1379	0.1256	0.1140	0.1011	0.0881	0.0810	0.0427
Newey-West S.E.	0.0080	0.0063	0.0068	0.0062	0.0064	0.0067	0.0071	0.0074	0.0076	0.0076	0.0077	0.0083
t-statistic	6.58	18.55	22.08	24.16	22.47	20.74	17.82	15.34	13.23	11.64	10.56	5.12
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
R²	0.0028	0.0203	0.0269	0.0320	0.0286	0.0256	0.0192	0.0149	0.0111	0.0085	0.0072	0.0017
Adjusted R²	0.0027	0.0202	0.0268	0.0319	0.0286	0.0256	0.0192	0.0148	0.0111	0.0084	0.0071	0.0016
Proportion of the effect explained by mood	48.2%	79.1%	87.4%	86.9%	85.4%	84.0%	77.9%	73.2%	68.7%	62.3%	60.6%	39.7%
Panel D: Weighted average
λ₀	-0.3175	-0.7209	-0.9259	-0.9199	-0.8829	-0.8471	-0.7691	-0.6976	-0.6174	-0.5362	-0.4927	-0.2559
Newey-West S.E.	0.0510	0.0423	0.0453	0.0418	0.0430	0.0446	0.0468	0.0491	0.0503	0.0496	0.0501	0.0529
t-statistic	-6.23	-17.04	-20.44	-22.01	-20.53	-18.99	-16.43	-14.21	-12.27	-10.81	-9.83	-4.84
p-value	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%
λ₁	0.0512	0.1162	0.1493	0.1483	0.1424	0.1366	0.1240	0.1125	0.0996	0.0865	0.0794	0.0413
Newey-West S.E.	0.0080	0.0063	0.0068	0.0062	0.0064	0.0066	0.0070	0.0074	0.0076	0.0076	0.0077	0.0083
t-statistic	6.42	18.39	22.09	24.07	22.32	20.60	17.64	15.18	13.07	11.44	10.37	4.98
p-value	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
R²	0.0027	0.0200	0.0270	0.0318	0.0284	0.0253	0.0189	0.0146	0.0109	0.0083	0.0070	0.0016
Adjusted R²	0.0026	0.0199	0.0269	0.0318	0.0283	0.0253	0.0189	0.0146	0.0108	0.0082	0.0069	0.0015
Proportion of the effect explained by mood	45.9%	77.9%	87.6%	86.5%	84.7%	83.0%	76.6%	71.8%	67.1%	60.5%	58.8%	37.3%

Table 3 provides analysis of the day-of-the-week effect with mood as the explanatory variable. The results are reported in four panels corresponding to four types of mood scores presented in Table 1 (columns 1, 3, 4 and 5 in Table 1). The regression coefficient of the mood variable is positive and statistically significant for all deciles and portfolios suggesting that the behavioral hypothesis is a possible explanation for the day of the week effect. The bottom line of each panel presents the proportion of the daily variation of average abnormal returns explained by mood, measured as the ratio of the R² of the mood variable regression to the R² of the respective dummy variable regression. Using the simple average mood template (Panel C), these proportions are approximately 79% and 48% for the EW and VW portfolios, respectively. The results for the R² ratios are declining with size suggesting that the ability of mood to explain the day-of-the-week effect declines with market capitalization. All the tests are adjusted for heteroscedasticity and serial correlation using Newey-West standard errors

Back to article page