Detection of Outliers and Patches in Bilinear Time Series Models

Table 1

The simulated results of the model BL$(\mathrm{2,2},\mathrm{1,1})$ via adaptive Gibbs algorithm.

	${\beta }_{31}$	${\beta }_{32}$	${\beta }_{33}$	${\beta }_{34}$	${\beta }_{35}$	${\beta }_{50}$
True value	$5$	$-4$	$5$	$-4$	$3$	$-6\mathrm{\hspace{0.17em}\hspace{0.17em}}$
Posterior mean	$4.7139$	$-3.6529$	$4.1193$	$-6.9487$	$4.7747$	$-2.6025\mathrm{\hspace{0.17em}\hspace{0.17em}}$
Standard deviation	$0.6211$	$0.6555$	$0.6044$	$0.6374$	$0.6711$	$0.5813\mathrm{\hspace{0.17em}\hspace{0.17em}}$