Table 4.
Difference-in-Difference (DiD) estimation and panel data regression model of the impact of low-cost carrier on New Zealand's domestic tourism (January 2008–July 2015).
| Dependent variable |
ln(Domestic guest nights)it |
|
|---|---|---|
| DiD estimation |
Panel data regression model |
|
| Explanatory variables | Coefficients | Coefficients |
| Constant | -41.2378*** (-3.94) |
11.3305*** (210.46) |
| Trend | -0.0128*** (-3.09) |
– |
| Treat_regionit | 0.2517*** (5.39) |
– |
| Post_LCCit | 0.3429*** (3.28) |
– |
| Treat_regionit* Post_LCCit | 0.0001*** (28.71) |
– |
| LCC_dummyit | – | 0.0716** (2.16) |
| ln(GDP per capita)t | 3.9980*** (3.15) |
3.7379*** (3.64) |
| ln(RTI -Accommodation)it | 0.2578*** (2.81) |
0.1769*** (5.36) |
| ln(RTI -Food & Beverage)it | 0.4901*** (4.16) |
0.2485*** (3.28) |
| ln(Aviation fuel price)t | -0.1715 (-0.80) |
-0.0465 (-1.14) |
| ln(Petrol price)t | -0.0267 (-0.05) |
-0.1720 (-1.36) |
| Exchange rate (NZD vs.USD)t | 0.0207 (0.07) |
-0.3479** (-2.45) |
| Global financial crisis 2008/09 | -0.0325 (-0.49) |
-0.0096 (-0.35) |
| Christchurch earthquakes 2011 | 0.0086 (0.20) |
-0.0503*** (-4.68) |
| ln(HHI Index)it | 1.7944*** (8.14) |
0.2794*** (2.74) |
| Seasonal dummy(1) | – | 0.2414*** (4.65) |
| Seasonal dummy(2) | – | 0.0504 (1.26) |
| Seasonal dummy(3) | – | 0.0207 (0.37) |
| Seasonal dummy(4) | – | 0.0609 (1.36) |
| Seasonal dummy(5) | – | -0.1479** (-2.45) |
| Seasonal dummy(6) | – | -0.2403*** (-2.90) |
| Seasonal dummy(7) | – | -0.1096 (-1.18) |
| Seasonal dummy(8) | – | -0.1363 (-1.45) |
| Seasonal dummy(9) | – | -0.1265 (-1.45) |
| Seasonal dummy(10) | – | -0.0744 (-1.03) |
| Seasonal dummy(11) | – | -0.0777 (-1.33) |
| Adjusted R2 | 0.649 | 0.939 |
| Observations | 910 | 910 |
Remarks: ** and *** indicate that the explanatory variable is significant at the 0.05 and 0.01 significance level, respectively. t-statistics are printed in parentheses. The difference-in-difference estimation is based on the OLS estimation. The results of Hausman test verified that the panel data regression model favoured using the random-effect model.