论文检索

共同基金绩效的自助法分析AI 解读

We study bootstrap methods for fund performance evaluation. We first show that two prominent bootstrap tests have biased test sizes in a large cross-section with short time series and lack test power to detect skilled funds when a substantial number of unskilled funds are present. We then develop the theory for a valid bootstrap Hotelling’s T-squared test for zero alpha. We apply the proposed bootstrap test in a practical two-step procedure to identify skilled funds. Our…

利用日度基金收益检验市场择时能力AI 解读

Using daily mutual fund returns to estimate market timing, some econometric issues, including heteroscedasticity, correlated errors, and heavy tails, make the traditional least-squares estimate in Treynor–Mazuy and Henriksson–Merton models biased and severely distort the t-test size. Using ARMA-GARCH models, weighted least-squares estimate to ensure a normal limit, and random weighted bootstrap method to quantify uncertainty, we find more funds with positive timing ability…

一般非对称依赖的检验AI 解读

Summary We propose a modified mutual information measure to capture general asymmetric dependence between two random variables. Based on this measure, we propose a test of asymmetric dependence and examine its finite‐sample performance. We show that our test has better power than competing tests with alternative dependence measures. Using the new test, we find significant asymmetric dependence in returns of commonly used stock portfolios and the market return both in the US…