Hierarchical linear model (HLM) for exports during the pre- and post-era
| Year | Model | R2 | F(df) | P | R2 change | F(df) change | P |
|---|---|---|---|---|---|---|---|
| Pre-PCE era (2010,2012,2014) | 1 | 0.821 | 63.216(4,55) | 0.000 | |||
| 2 | 0.834 | 54.444(5,54) | 0.000 | 0.013 | 4.278(1,54) | 0.043** | |
| 3 | 0.835 | 44.659(6,53) | 0.000 | 0.000 | 0.128(1,53) | 0.722 | |
| 4 | 0.835 | 37.567(7,52) | 0.000 | 0.000 | 0.011(1,52) | 0.916 | |
| 5 | 0.842 | 34.066(8,51) | 0.000 | 0.007 | 2.413(1,51) | 0.126 | |
| 6 | 0.849 | 31.310(9,50) | 0.000 | 0.007 | 2.303(1,50) | 0.135 | |
| GDP (1.644)***, TRFR (25.335)***, LPCust (−1.112)**, N = 60 | |||||||
| Post-PCE era (2016,2018) | 1 | 0.668 | 17.625(4,35) | 0.000 | |||
| 2 | 0.751 | 20.506(5,34) | 0.000 | 0.083 | 11.295(1,34) | 0.002*** | |
| 3 | 0.751 | 16.614(6,33) | 0.000 | 0.000 | 0.043(1,33) | 0.838 | |
| 4 | 0.751 | 13.819(7,32) | 0.000 | 0.000 | 0.016(1,32) | 0.900 | |
| 5 | 0.754 | 11.890(8,31) | 0.000 | 0.003 | 0.350(1,31) | 0.558 | |
| 6 | 0.768 | 11.034(9,30) | 0.000 | 0.014 | 1.783(1,30) | 0.192 | |
| TRFR (15.481)***, IND (11.172)***, LPCust (−2.261)***, N = 40 | |||||||
| Pre-PCE era (OLPI) | 1 | 0.821 | 63.219(4,55) | 0.000 | |||
| 2 | 0.826 | 51.148(5,54) | 0.000 | 0.004 | 1.333(1,54) | 0.253 | |
| GDP (2.019)***, TRFR (25.91)***, N = 60 | |||||||
| Post-PCE era (OLPI) | 1 | 0.668 | 17.625(4,35) | 0.000 | |||
| 2 | 0.741 | 19.477(5,34) | 0.000 | 0.073 | 9.588(1,34) | 0.004 | |
| Year | Model | ||||||
|---|---|---|---|---|---|---|---|
| GDP (1.644)***, TRFR (25.335)***, LPCust (−1.112)**, | |||||||
| TRFR (15.481)***, IND (11.172)***, LPCust (−2.261)***, | |||||||
| GDP (2.019)***, TRFR (25.91)***, | |||||||
Note(s): GDP (1.306)***, TRFR (15.733)**, IND (10.870)***, OLPI (−2.312)***, N = 40
***p < 0.01, **p < 0.05, *p < 0.1. Model 1 uses the general OLPI index (overall); model 2 uses the LPI customs index; model 3 uses the LPI infrastructure; model 4 uses the LPI quality of logistics services index; Model 5 uses the LPI tracking and tracing index; model 6 uses the LPI timeliness. Coefficients are based on log-transformed variables and therefore reflect scaled associations rather than standardized beta coefficients
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.