My WebLink
|
Help
|
About
|
Sign Out
Home
Browse
Search
FLOOD04787
CWCB
>
Floodplain Documents
>
Backfile
>
4001-5000
>
FLOOD04787
Metadata
Thumbnails
Annotations
Entry Properties
Last modified
1/25/2010 6:47:15 PM
Creation date
10/5/2006 12:59:32 AM
Metadata
Fields
Template:
Floodplain Documents
County
Statewide
Basin
Statewide
Title
Colorado Association of Storm Water and Floodplain Managers 8th annual Conference
Date
9/22/1997
Prepared For
State of Colorado
Prepared By
CASFM
Floodplain - Doc Type
Educational/Technical/Reference Information
There are no annotations on this page.
Document management portal powered by Laserfiche WebLink 9 © 1998-2015
Laserfiche.
All rights reserved.
/
181
PDF
Print
Pages to print
Enter page numbers and/or page ranges separated by commas. For example, 1,3,5-12.
After downloading, print the document using a PDF reader (e.g. Adobe Reader).
Show annotations
View images
View plain text
<br />. The dependent variable in the regressions is the assessed value of the properties at the <br /> <br />time of sale. It is not significantly different from the actual sale value. This is not <br /> <br />surprising because with California's Proposition 13, assessed property values are updated to <br /> <br />the new market price at the time of sale. A Box-Cox transformation was employed to <br /> <br />determine the functional form which fits the data and the variables which best explained the <br /> <br /> <br />variation in the dependent variable (the assessed value in real terms). <br /> <br />To conduct multiple regression analysis, the question of functional form was <br /> <br /> <br />addressed using the Box-Cox transformation method (Greene 1992). This method allows for <br /> <br /> <br />the transformation of both the dependent and independent variables. The Box-Cox <br /> <br /> <br />transformation indicated a non-linear functional form fit the data. This model was used in <br /> <br /> <br />the multiple regression analysis. The Box-Cox transformation method allows for <br /> <br />. transformations of both the 1eft-hand-side and right-hand-side variables in a regression <br /> <br />equation, depending on the nature of the data and the interaction of the independent and <br /> <br /> <br />dependent variables. Preliminary regressions were attempted using all four possible Box-Cox <br />specifications: transformation of the dependent variable by lambda, the independent variables <br />by lambda, both independent and dependent variables by lambda and transforming the <br />dependent variable by theta and the independent variables by lambda. Using LIMDEP <br />(Greene 1992), the model chosen for this data set transforms only the dependent variable, <br />assessed property value. Based on the minimum value of the log-likelihood value, <br />preliminary regressions indicated this transformation was best. Lambda represents the <br />exponent to which the dependent variable is raised. The model is nonlinear as is suggested <br />by the hedonic price theory reviewed previously, <br /> <br />. <br /> <br />11 <br />
The URL can be used to link to this page
Your browser does not support the video tag.