Passenger_Movement_Forecast_at_Delhi_International_Airport

Term paper Forecasting of Aircraft Movements: Delhi Airport. Submitted T*o* Submitted By Objective The objective of the study is to develop an econometric model by which Delhi airport operator can forecast the number of aircraft movement in a particular month of the year, given a certain number of passenger and tones of freight. By knowing the number of aircraft movement an airport operator can take decision for potential requirements such as number and size of runway required, whether a new terminal building is required or not, number of taxiways, bays, gates, etc. it will also help operator in its strategic planning and organizational planning. Methodology To develop the econometrics model data has been collected on the three variables such as; No. of Aircraft Movement Monthly No. of passenger Movement Monthly Tonnes of freight Movement monthly Data is collected from year April 2005 to May 2008 Multivariate regression line has been developed with the help of MINITAB and then Aircraft Movement for the period of Dec. 2009 to April 2010. And study also try to check that how number of Aircraft Movements depend on the number of passenger movement and tonnes of freight movement. Study also deals with the problem of Heteroscedasticity and multicollinearity Finding and Analysis: Data has been collected from the DGCA from the period of April 2005 to May 2008 {draw:frame} {draw:frame} {draw:frame} It is clear from the above graphs that Aircraft Movement, passenger movement and Tonnes of freight have upward trend. *Regression Analysis: Aircraft movement versus No. of Pax*, Freights in The regression equation is Predictor Coef SE Coef T P Constant -53 1067 -0.05 0.960 No. of Pax 0.0070424 0.0003381 20.83 0.000 Freights in Tonne 0.10771 0.03783 2.85 0.008 S = 509.840 R-Sq = 95.7% R-Sq (adj) = 95.4% Analysis of Variance Source DF SS MS F P Regression 2 190397447 95198724 366.24 0.000 Residual Error 33 8577907 259937 Total 35 198975355 Source DF Seq SS No. of Pax 1 188290737 Freights in Tonne 1 2106710 Interpretations: R square is 95.7% which means95.7% variation in Number of Aircraft movement can be explained or depend on the number of passenger movement and freight movements. For both independent variables are significant as T stat value for paxvariable is 20.83 and freight is 2.85 Test of Multicollinearity As R square value is 95.7%, there may be multicollinearity exists, *Correlations: No. of Pax*, Freights in Tonne Pearson correlation of No. of Pax and Freights in Tonne = 0.554 P-Value = 0.000 VIF= (1/1-Ri) = 2.22 which means low level of multicollinearity. SE of coefficients: Predictor SE Coef T P Constant 1067 -0.05 0.960 No. of Pax 0.0003381 20.83 0.000 Freights in Tonne 0.03783 2.85 0.008 As standard error of constant is high and the t stat is not significant which indicates the existence of multicollinearity. Other variable such as No. of passenger and freight are low, and t value shows both variables are significant. Dropping of some observation: *Regression Analysis: Aircraft mov*. Versus No. of *Pax*, Freights in The regression equation is Aircraft movement = - 596 + 0.00708 No. of Pax + 0.123 Freights in Tonne Predictor Coef SE Coef T P Constant -596 1192 -0.50 0.621 No. of Pax 0.0070757 0.0003628 19.50 0.000 Freights in Tonne 0.12293 0.04078 3.01 0.005 S = 520.039 R-Sq = 95.1% R-Sq(adj) = 94.7% Analysis of Variance Source DF SS MS F P Regression 2 156670478 78335239 289.66 0.000 Residual Error 30 8113225 270441 Total 32 164783703 Theils measure Excluding freight movement: The regression equation is Aircraft movement = 2630 + 0.00760 No. of Pax S = 583.946 R-Sq = 93.6% R-Sq(adj) = 93.4% Analysis of Variance Source DF SS MS F P Regression 1 154212905 154212905 452.25 0.000 Residual Error 31 10570798 340993 Total 32 164783703 Excluding Passenger movement: Regression Analysis: Aircraft movement versus Freights in Tonne The regression equation is Aircraft movement = - 1762 + 0.505 Freights in Tonne Predictor Coef SE Coef T P Constant -1762 4332 -0.41 0.687 Freights in Tonne 0.5047 0.1302 3.88 0.001 S = 1891.97 R-Sq = 32.7% R-Sq (adj) = 30.5% Analysis of Variance Source DF SS MS F P Regression 1 53817854 53817854 15.03 0.001 Residual Error 31 110965849 3579544 Total 32 164783703 _M=R2-{(R2-r21) + (R2-r22)}_ M=95.7-{(95.7-93.6) + (95.7-32.7)} =30.6 It means multicollinearity exists. Conclusion: Test of Heteroscedasticity By ordinary least square graph method: {draw:frame} {draw:frame} By looking at the graph it can be concluded that there is Heteroscedasticity in the data. {draw:frame} {draw:frame} Hypothesis Testing: *H0*: Error variances are equal *H1*: error variances are not equal Here we see that calculated value for F is less than tabulated value hence we accept the null hypothesis. Test of Autocorrelation: Durbin-Watson statistic *Regression Analysis: Aircraft mov* versus No. of *Pax*, Freights in The regression equation is Aircraft movement = - 53 + 0.00704 No. of Pax + 0.108 Freights in Tonne S = 509.840 R-Sq = 95.7% R-Sq(adj) = 95.4% Analysis of Variance Durbin-Watson statistic = 1.20663 It indicates there is existence of Autocorrelation in the data given. Trend modeling Regression Analysis: Aircraft movement versus Period The regression equation is Aircraft movement = 11186 + 221 Period Predictor Coef SE Coef T P Constant 11186.1 177.5 63.03 0.000 Period 220.992 8.365 26.42 0.000 S = 521.367 R-Sq = 95.4% R-Sq(adj) = 95.2% Analysis of Variance Source DF SS MS F P Regression 1 189733362 189733362 698.00 0.000 Residual Error 34 9241992 271823 Total 35 198975355 Forecasted Aircraft Movements with the help of Linear Trend Model