Modeling of 2D diffusion processes based on microscopy data: parameter estimation and practical identifiability analysis

Table 1 Estimation results

Name	value	MLE	parameter bounds		confidence intervals		identifiability
	θ	$\hat{θ}$	θ _min	θ _max	C _i,min	C _i,max
D	0.5	0.4985	0.01	10	0.4939	0.5044	practical identifiable
α	0.1	0.0995	0.01	10	0.0969	0.1059	practical identifiable
k ₁	5	5.0375	0.01	10	4.6841	5.1761	practical identifiable
k ₋₁	1	1.0281	0.01	10	0.9704	1.0687	practical identifiable
γ	0.1	0.0669	0.01	10	< 0.0100	0.3272	practically non-identifiable

True value of each estimated parameter compared to maximum likelihood estimate for a local multiple start optimization. Bounds for each parameter where used for estimation.

ISSN: 1471-2105