Laserfiche WebLink
RE7.A. - Page 241 Expansion Project#:11637.0 <br /> Auyus<<v,LV1L Page 29 <br /> Figure 11. Regression Analysis of Trip Generation Versus Fuel Sales <br /> 600 <br /> � <br /> O <br /> � <br /> � 500 ■ ■ <br /> a� <br /> � ■ • ■ <br /> a� <br /> C7 ■ ■■ ■ <br /> � 400 � <br /> � • ■ <br /> ■ <br /> 3 • <br /> = 300 <br /> Y <br /> fC <br /> � 20o Linear Regression Trend Line Logarithmic Regression Trend Line <br /> a y= 0.0043x+ 273.59 y= 143.121n(x) - 1070.9 <br /> � Rz = 0.1762 Rz = 0.1899 � <br /> � <br /> Y 1OO <br /> a� <br /> a� <br /> � <br /> 0 <br /> Average Daily Gallons Sold <br /> Both a linear and logarithmic regression analyses were conducted on the trip generation data for <br /> both independent variables (number of fueling positions and fuel volume sold). As shown in the <br /> Figure 10, the RZ values for the regression analysis for trip generation against fueling positions were <br /> both less than 0.035 and, as shown in Figure 11, the RZ values for the regression analysis for trip <br /> generation against gallons sold were both less than 0.19. The RZ value is a statistical value (called the <br /> coefficient of determination) that is used by the ITE Trip Generation manual in the context of <br /> statistical models whose main purpose is to predict future outcomes on the basis of other related <br /> information. It is the proportion of variability in a data set that is accounted for by the statistical <br /> model and provides a measure of how well future outcomes are likely to be predicted by the model. <br /> An RZ of 1.0 indicates that the regression line (or model) perfectly fits the data and an RZ value of 0 <br /> indicates there is no relationship between the data and a given independent variable. The value RZ = <br /> 0.70 may be interpreted as indicating that the model fits 70% of the data and that for 70% of <br /> situations the model will reasonably predict the future outcome. The remaining 30% can be <br /> explained by unknown or inherent variability. The ITE Trip Generation manual also presents linear <br /> and logarithmic regression analysis for the trip generation data it contains. A typical rule of thumb <br /> when applying the Trip Generation manual is that data where the RZ value is less than 0.75 does not <br /> show a strong or reasonably predictive correlation or trend in the outcome results. In other words, if <br /> Kittelson&Associates,Inc. Boise,Idaho <br />