Table 6 Hypothesis testing results for the proportion of times the average return of Type-1 is higher than Type-3 (\(p\)-values < 0.05), rejecting the null hypothesis in favor of the alternative). In the round brackets, the proportion of the times Type-1 outperforms Type-3 in average return is indicated. The symbols in the first column denote the holding period (T), while those in the first row represent the portfolio size, where \(m = 5, 10, 20, \text { and } 30\), respectively. The table reports results for portfolios constructed using the Equal Weight (\(\textbf{w}^{\textrm{EW}}\)) and Markowitz (\({w}^\mathrm{{MW}}\)) weighting schemes.
From: A novel portfolio construction strategy based on the core- periphery profile of stocks
T | 5 stocks | 10 stocks | 20 stocks | 30 stocks | ||||
|---|---|---|---|---|---|---|---|---|
\(\textbf{w}^{\textrm{EW}}\) | \(\textbf{w}^{\textrm{MW}}\) | \(\textbf{w}^{\textrm{EW}}\) | \(\textbf{w}^{\textrm{MW}}\) | \(\textbf{w}^{\textrm{EW}}\) | \(\textbf{w}^{\textrm{MW}}\) | \(\textbf{w}^{\textrm{EW}}\) | \(\textbf{w}^{\textrm{MW}}\) | |
50 | 0.99 (0) | 64.2 (0.99) | 0.99 (0) | 71.8 (0.1) | 0.99 (0) | 0.84 (0) | 0.99 (0) | 0.81 (0) |
60 | 1.0 (0) | 63.7 (0.99) | 1.0 (0) | 0.67 (0.97) | 0.99 (0) | 0.81 (0) | 0.99 (0) | 83.5 (0) |
70 | 1.0 (0) | 59.1 (1.0) | 1.0 (0) | 71.5 (0.15) | 0.99 (0) | 81.4 (0) | 0.99 (0) | 81.9 (0) |
80 | 1.0 (0) | 53.7 (1.0) | 1.0 (0) | 0.72 (0.04) | 1.0 (0) | 0.80 (0) | 0.99 (0) | 0.77 (0) |
90 | 1.0 (0) | 0.58 (1.0) | 0.99 (0) | 0.77 (0) | 0.99 (0) | 0.83 (0) | 0.99 (0) | 0.84 (0) |
100 | 1.0 (0) | 0.68 (0.79) | 0.99 (0) | 0.84 (0) | 0.99 (0) | 0.88 (0) | 0.99 (0) | 0.90 (0) |
110 | 1.0 (0) | 0.63 (0.99) | 0.99 (0) | 0.80 (0) | 0.99 (0) | 0.88 (0) | 0.99 (0) | 0.88 (0) |
125 | 1.0 (0) | 0.74 (0.005) | 0.99 (0) | 0.82 (0) | 0.99 (0) | 0.91 (0) | 0.98 (0) | 0.89 (0) |