Pooled Time Series Interpretation - Statswork

## Pooled Time Series Cross-Sectional Regression Analysis

### Interpretation

Pooled time series cross sectional analysis helps in offering explanations of the past, while simultaneously predicting the future behaviour of exogenous variables in relation to endogenous variables. Pooled time series cross sectional analysis allows the researcher to focus on more than one case in predicting social phenomenon.

Table 1 shows the results for the regression equation of (all hotels) measuring the impact of occupancy rate, interest, FAR and Tax rate on number of hotels from 1978 through the 2013. The table shows an adjusted R2 of 0.66, suggesting that 61 percent of the variance is being explained by the equation. The F value of 70.5 with a p<0.001 accepted at the .001 level of significance, suggests that the model is valid. The occupancy rate was negatively affected with the interest rate variable was negatively (β=-0.968, (SE: 2.203); p<0.001) affected with the number of hotels, while FAR (dummy variable) cost (β=11.80 (SE: 2.284), p<0.001) was positively affected the growth at p value <0.001, suggesting, that increase in interest rate there is a decrease in number hotels and while increase in FAR cost, there is an increase in number of hotels. Although the occupancy was negatively affected with the number of star hotels it did not reach statistically significance. There is a borderline significance between tax variable and number of hotels. In regards to tax variable, borderline significance was obtained with number of hotels.

Table 2 presents the results for the regression equation of 2 star hotels measuring the impact of occupancy rate, interest, FAR and Tax rate on number of hotels from 1978 through the 2013. The table shows an adjusted R2 of 0.86, suggesting that 86 percent of the variance is being explained by the equation. The F value of 44.03 with a p<0.001 accepted at the .001 level of significance, suggests that the model is valid. The interest rate variable was negatively affected with the number of hotels, while FAR (dummy variable) cost (β=19.609 (SE: 4.035), p<0.001) was positively affected the growth at p value <0.001, suggesting, that increase in land cost there is a decrease in number hotels and while decrease in FAR cost, there is an increase in number of hotels. Although the Interest rate was negatively affected with the number of 1-2 star hotels it did not reach statistically significant.

Table 3 shows the results for the regression equation of 3-4 star hotels measuring the impact of occupancy rate, interest, FAR and Tax rate on number of hotels from 1978 through the 2013. The table shows an adjusted R2 of 0.917, suggesting that 92 percent of the variance is being explained by the equation. The F value of 77.832 with a p<0.001 accepted at the .001 level of significance, suggests that the model is valid. The occupancy rate (β=-0.635;p<0.001) variable was negatively affected with the number of hotels, while land cost (β=0.00005 (SE: 4.035), p<0.001) was positively affected the hotels growth at p value <0.001, suggesting, that increase in land cost there is a increase in number of hotels and while increase in FAR cost, there is an increase in number of hotels but this did not reach statistically significant. In this case, Tax variable (dummy variable) was negatively (β=-13.139; p<0.0001) associated with the number of hotels. Although the Interest rate was negatively affected with the number of 3 star hotels it did not reach statistically significant.

The results for the regression equation of 3-4 star hotels measuring the impact of occupancy rate, interest, FAR and Tax rate on number of hotels from 1978 through the 2013 are presented in Table 4. The table shows an adjusted R2 of 0.968, suggesting that 96 percent of the variance is being explained by the equation. The F value of 208.5 with a p<0.001 accepted at the .001 level of significance, suggests that the model is valid. The occupancy rate (β=--0.199; p<0.001) variable was negatively affected with the number of hotels, while land cost (β=0.00001; p<0.001) was positively affected the hotels growth at p value <0.001, suggesting, that increase in land cost there is a increase in number of hotels and while increase in FAR cost, there is an increase in number of hotels but this did not reach statistically significant. In this case, Tax variable (dummy variable) was negatively associated with the number of hotels but did not reach statically significant.

The results for the regression equation of 5 star hotels measuring the impact of occupancy rate, interest, FAR and Tax rate on number of hotels from 1978 through the 2013 are presented in Table 5. The table shows an adjusted R2 of 0.936, suggesting that 94 percent of the variance is being explained by the equation. The F value of 104.0 with a p<0.001 accepted at the .001 level of significance, suggests that the model is valid. The land cost (β=--0.0005; p<0.001) variable was positively affected with the number of hotels. Suggesting that increase in land cost there is an increase in number of hotels.

The results for the regression equation of 5D star hotels measuring the impact of occupancy rate, interest, FAR and Tax rate on number of hotels from 1978 through the 2013 are presented in Table 6. The table shows an adjusted R2 of 0.824, suggesting that 82 percent of the variance is being explained by the equation. The F value of 33.877 with a p<0.001 accepted at the .001 level of significance, suggests that the model is valid. The occupancy rate (β=-0.324; p<0.001) variable was positively affected with the number of hotels. Suggesting that increase in occupancy cost there is an increase in number of hotels significantly.