Tracking market and non-traditional sources of risks in procyclical and countercyclical hedge fund strategies under extreme scenarios: a nonlinear VAR approach

Financial Innovation

Table 2 Descriptive statistics of the time-varying betas of hedge fund strategies

	Mean	Median	Max	Min	St-dev	Sharpe	ρ
General index	0.378	0.389	0.532	0.203	0.078	4.846	0.25***
Convertibles	0.246	0.236	0.560	0.009	0.115	2.139	0.57***
Distressed securities	0.376	0.397	0.579	0.099	0.113	3.327	0.47***
Equity market neutral	0.145	0.156	0.219	0.023	0.047	3.085	0.27***
Event driven	0.439	0.460	0.651	0.185	0.113	3.885	0.29***
Fixed income	0.158	0.129	0.413	−0.098	0.114	1.386	0.41***
Futures	0.004	−0.007	0.245	−0.137	0.078	0.051	0.01
Growth	0.844	0.890	1.143	0.418	0.187	4.513	0.19**
Long-short credit	0.129	0.118	0.314	−0.109	0.085	1.518	0.39***
Macro	0.110	0.112	0.190	0.023	0.035	3.143	0.04
Mergers	0.136	0.141	0.221	0.029	0.048	2.833	0.23***
Multistrategy	0.401	0.429	0.564	0.165	0.099	4.051	0.24***
Opportunity index	0.470	0.481	0.711	0.268	0.096	4.896	0.20**
Short-sellers	−0.976	−0.949	−0.264	−1.922	0.351	−2.781	0.11*
Value index	0.591	0.610	0.821	0.298	0.131	4.511	0.18***

The time-varying beta of strategy i is equal to: \(\beta_{it} = \frac{{{\text{cov}}_{t} \left( {R_{i} ,R_{m} } \right)}}{{{\text{var}}_{t} \left( {R_{m} } \right)}}\), where R_i is the return of strategy i and R_m is the market return, as measured by the S&P500’s return. The covariance and variance measures are computed with the MGARCH algorithm (Bollerslev et al. 1988; Engle and Colacito 2006; Engle 2016). The ρ coefficient associated with a strategy is equal to the coefficient of autocorrelation of the first-order degree of its return. The significance levels of the p-values are 1% indicated by ***; 5%, **; and 10%, *.