The Hesitation of Anxious Traders in an Agent-Based Model

Table 4

Simulations with various degrees of risk aversion. Note: The agents’ benefits and momentum strengths for various degrees of risk aversion in agent evolutions are presented in this table. The indicator of hesitation (represented by d) lies between 1 and 5. When d equals 1, the expectation for price of the anxious agent can be expressed as . Hence, the anxious agent does not embody hesitation since the agent only relies on the single-period forecast error in such a case. However, hesitation begins to emerge if d is 2 or over 2 since the agent refers to multiperiod forecast errors. The risk aversion (RA) of the anxious agent lies between 0.1 and 5. In the pursuit of accuracy regarding statistical inference, we follow Gatti and Grazzini [88] and García-Magariño et al. [89] to perform a one-sample t-test for each window. This procedure enables us to calculate the averages of the t-statistics across windows. The upper bound of the evolutionary mechanism is 0.1.

	Mean					t (mean)
	RA = 0.1	RA = 0.5	RA = 1	RA = 2	RA = 5	RA = 0.1	RA = 0.5	RA = 1	RA = 2	RA = 5
Panel A: anxious agents’ benefit minus fundamentalist agents’ benefit
d = 1	0.124	0.085	0.086	0.086	0.086	4.791	4.935	5.410	5.574	5.733
d = 2	0.133	0.077	0.083	0.083	0.086	3.581	4.627	5.330	5.461	5.688
d = 3	0.498	0.074	0.083	0.084	0.086	3.974	4.443	5.309	5.483	5.725
d = 4	2.200	0.066	0.080	0.084	0.085	3.852	3.880	5.080	5.487	5.626
d = 5	2.376	0.063	0.082	0.083	0.086	3.464	3.687	5.159	5.405	5.700
Panel B: momentum strength
d = 1	−0.349	−0.269	−0.292	−0.300	−0.304	−5.177	−4.182	−4.647	−4.852	−4.959
d = 2	−0.165	−0.223	−0.222	−0.224	−0.214	−1.857	−3.723	−3.660	−3.682	−3.531
d = 3	−0.102	−0.142	−0.155	−0.186	−0.193	−0.974	−1.979	−2.160	−2.571	−2.680
d = 4	0.034	−0.220	−0.264	−0.249	−0.232	0.134	−2.933	−3.671	−3.454	−3.233
d = 5	0.002	−0.132	−0.102	−0.127	−0.161	0.039	−1.673	−1.362	−1.718	−2.207