Laserfiche WebLink
<br />. coefficients on education trail, stabilize stream banks, fish habitat, acquire land and reduce <br /> <br />flood damage, which are each significant at the five percent level in their respective <br /> <br />regressions. While the year of sale is not significant at the ten percent level, it has a t- <br /> <br />statistic of greater than one and is included in the model to account for possible increases in <br /> <br /> <br />real property prices over the time frame of the study. <br />Table II shows regression coefficients for the models with two measures which are <br />grouped together in a restoration package. Both reduce flood damage and stabilize stream <br />banks are by themselves positive and significant. As shown in Table n, stabilize stream <br />banks and education trail are both significant when included together. Both reduce flood <br /> <br />damage and education trail, however, were not significant when included together in the <br /> <br />model. <br /> <br />. <br /> <br />Property Value Results <br />The individual values for the stream improvement measures can be calculated from <br />the coefficients estimated by the model. The method evaluates the change in property price <br />due to the measure itself, holding all other. variables constant at zero. The following <br />equation from Greene (1992) for the Box-Cox nonlinear regression model was used to <br />calculate the individual values. <br /> <br />(3) y=[<^*,8X)+1]II), <br /> <br />As an example of this calculation, the value of the fish habitat variable is set equal to one. <br /> <br /> <br />This indicates the completion of that measure. If the other variables are held constant at <br /> <br />zero, the equation becomes: <br /> <br />. <br /> <br />13 <br />