Table 4. Summary of posterior distributions for Model 3: City-level and district-level effects

Parameter	Mean	SD	5% HDI	95% HDI	MCSE mean	MCSE SD	R-hat
α[Seoul]	0.0082	0.0055	–0.0005	0.0169	0.0002	0.0002	1.00
α[Busan]	0.0016	0.0051	–0.0061	0.0103	0.0001	0.0001	1.00
α[Daegu]	0.0017	0.0057	–0.0069	0.0114	0.0002	0.0001	1.01
α[Incheon]	0.0012	0.0054	–0.0077	0.0101	0.0001	0.0001	1.01
α[Gwangju]	–0.0029	0.0059	–0.0128	0.0062	0.0002	0.0001	1.00
α[Daejeon]	–0.0051	0.0065	–0.0156	0.0044	0.0002	0.0002	1.00
α[Ulsan]	–0.0045	0.0063	–0.0145	0.0051	0.0002	0.0002	1.00
β0 (intercept)	0.0101	0.0002	0.0098	0.0103	0.0000	0.0000	1.00
β1 (national-level momentum effect)	–0.4388	0.0043	–0.4456	–0.4321	0.0001	0.0001	1.00
σ (residual SD)	0.1059	0.0001	0.1057	0.1061	0.0000	0.0000	1.00
σα (city-level momentum effect SD)	0.0081	0.0049	0.0006	0.0144	0.0002	0.0001	1.01
σδ (district-level momentum effect SD)	0.0123	0.0024	0.0087	0.0164	0.0001	0.0001	1.01

Note: This table reports the posterior summaries for the Model 3. The posterior mean and SD represent the central tendency and dispersion of the estimated parameters. The 5% and 95% highest density intervals (HDIs) define the central 90% credible region for each parameter. Monte Carlo standard errors (MCSE) assess the numerical accuracy of the posterior mean and SD estimates. R-hat values close to 1 indicate satisfactory convergence across Markov chain Monte Carlo (MCMC) chains. The parameter α[city] denotes the city-level momentum effect. District-level momentum effects (δ_k) are not reported here due to the limited space.