Table 4.

The rows in the table correspond (respectively) to the means of point estimates, Monte Carlo empirical standard deviation, means of estimated asymptotic standard deviation, true asymptotic standard deviation and actual coverage of interval estimates (using the estimated asymptotic standard deviation) for the parameter θ₀=1 in the linear ordinary differential equation case (Equation (16)) (initial value ξ₀=1 was estimated as well). The results are based on 1000 Monte Carlo simulations. The observed time points are equidistant on [0,10], and the errors are normal with zero expectation and σ as in the table. The one‐step point estimates are given by Equation 6; the interval estimates are defined in Equation 15

n = 250			n = 500			n = 1000
σ=0.1	σ=0.2	σ=0.3	σ=0.1	σ=0.2	σ=0.3	σ=0.1	σ=0.2	σ=0.3
1.0010	1.0010	0.9990	1.0000	1.0000	1.0010	1.0000	1.0000	1.0000
0.0130	0.0260	0.0400	0.0090	0.0190	0.0280	0.0070	0.0130	0.0200
0.0130	0.0280	0.0390	0.0090	0.0180	0.0290	0.0070	0.0130	0.0200
0.0130	0.0270	0.0400	0.0100	0.0190	0.0290	0.0070	0.0130	0.0200
0.9460	0.9540	0.9610	0.9540	0.9520	0.9530	0.9430	0.9550	0.9470