Technical Report - Laserfiche WebLink

Home

The second step in the process of projecting registered aircraft was of a judgmental nature. Each of the projections was examined in the context of possible impacts of improved facilities and service, as well as the relationship between regional and national trends. Consideration of these factors, combined with previous experience concerning reliever system projections, led to a subjective determination as to the validity of various projections. The selected projections, as a result, contain all the desired features of objective and subjective analyses. Of the four independent projection methodologies, two utilize one basic analytical tool, regression analysis. The term regression analysis refers to the statistical technique by which estimates are made of the values of dependent variables based on knowledge of the values of one or more other variables, called independent variables. For the socioeconomic regression in this analysis, the. dependent variable was the appropriate airport activity figure, such as registered aircraft, while the independent variable • was a socioeconomic factor, such as population. The historical trend analysis fitted historical data to classical growth curves and extended the demand element into future periods by regression analysis. The procedure for this analysis resembled the socioeconomic regression, except that time was substituted as the independent variable. The most common growth curves are in the form of linear, exponential, or logarithmic equations. For this study, the equational form of the projection line most often developed was lineaz. To measure the relationship between dependent and independent variables in both the regression and the trend analysis, the coefficient of determination (R~ and the standazd error of the estimate (y) were used. The R' statistic indicates whether the given independent variable might be related to the dependent variable. It does so by quantifying the percentage of the dependent variable's variation about its arithmetic mean that is explained by the model. For instance, an R' of 0.80 indicates that 80 percent of the variance that existed around the dependent variable's mean has now been explained by examining the variance around the regression line (i.e., only 20 percent of the variance that once existed still exists). The socioeconomic regression and historical trend analysis II-3