The value premium and uncertainty: An approach by support vector regression algorithm

Figures & data

Table 1. Variable description

Name	Variable Description
$r_{ft}$	1-Year Government Bond Yield
$r_{Mt}$	The return rate of the VN-index portfolios
$B{M}_{it}$	Book value over the market value of portfolio i
$B{M}_{Mt}$	Book value over the market value of the market portfolio
$I_{it}$	Uncertainty index. Iit = 0 implies that the portfolio is stable relative to the market.
$p_{it}$	Weighted rate of return
$p_{it}-r_{mt}$	Outstanding rate of return of the portfolio
$B{M}_{it}-B{M}_{Mt}$	Book-to-market ratios in excess of market

Figure 1. The VN-index point (divided by 1000) and the BM_M ratio for the market portfolio.

Table 2. Descriptive statistics returns in excess of the market

Portfolio	P1	P2	P3	P4	P5	P6
Mean	0.540	0.402	0.316	−0.348	−0.375	−0.320
Median	0.260	0.564	0.731	−0.315	−0.269	−0.520
St. Dev.	2.791	2.133	2.076	1.928	1.879	1.351
Kurtosis	0.379	0.748	−0.083	0.485	0.254	0.365
Skewness	−0.240	−0.228	−0.020	−0.101	−0.061	−0.032
Minimum	−12.862	−12.546	−11.546	−11.431	−13.004	−12.382
Maximum	7.378	10.399	9.950	10.964	9.126	12.381

Table 3. Correlation between portfolios

	P1	P2	P3	P4	P5	P6
P1	1	0.925	0.785	−0.320	−0.242	−0.318
P2		1	0.877	−0.531	−0.315	−0.168
P3			1	−0.423	−0.445	−0.287
P4				1	0.893	0.773
P5					1	0.930
P6						1
Cor(P, rMt)	0.751	0.785	0.884	0.896	0.832	0.771

Table 4. Regression of return premium on market excess BM ratio

Portfolio	Coefficients		P value		R^2	Durbin–Watson
	α	β	α	β
P1	-1.231	2.395	0.002	0.001	0.421	1.41
P2	-1.516	2.547	0.007	0.001	0.518	1.35
P3	-1.486	1.983	0.011	0.001	0.532	1.48
P4	0.355	0.798	0.121	0.069	0.328	1.55
P5	0.218	1.116	0.218	0.095	0.211	1.47
P6	0.162	0.129	0.351	0.522	0.175	1.68

Table 5. Regression results of model 2

Portfolio	Coefficients				R^2	Adj. R^2	Durbin—Watson
	β0	β1	β2	β3
P1	−0.511**	−0.241	1.521***	0.873*	0.643	0.614	2.01
P2	−0.328**	−0.612*	1.438***	0.629*	0.724	0.639	1.94
P3	−0.271**	−0.386*	1.332***	1.125*	0.612	0.593	2.11
P4	0.167*	−0.218	0.327**	0.763*	0.524	0.469	2.07
P5	0.217*	−0.071	0.375**	0.213*	0.596	0.541	1.92
P6	0.087	0.116*	−0.105*	1.141	0.538	0.493	1.89

Notes: (Significant level: 0.01“***”, 0.05“**”, 0.1“*”).

Table 6. The deviations of 3 models

Model	Count	Sum	Average	Variance
Model 1	324	917.6	2.83	0.23
Model 2	324	753.12	2.32	0.39
Model 3	324	583.1	1.80	0.43

Table 7. ANOVA result

Source of Variation	SS	df	MS	F	P-value
Between Groups	172.69	2	86.34	232.88	0.0001
Within Groups	359.26	969	0.37
Total	531.95	971

Figure 2. Deviation distribution in three models.