Data Science Tools: R, Excel, KNIME, & OpenOffice by Christopher Greco

Author:Christopher Greco [Greco, Christopher]
Language: eng
Format: epub, mobi
Publisher: Mercury Learning and Information
Published: 2020-08-15T04:00:00+00:00

The result of the regression analysis is as follows. Please notice all the different numbers on this screen. The ones that are important to the analyst immediately will be those that encompass the regression equation.

The numbers that are of interest to the analyst include the ones that are in the column “Coefficients” for both “Intercept” and “X Variable” (which in this case is TOR_LENGTH). The resulting regression equation will be y=0.0164x + 4.22962. This means that if the analyst wants to know what tornado width they should expect, the analyst places the tornado length for the “x” variable, multiplies it by .0164, and adds 4.33962 to find the approximate tornado width. Here is where regression can be misused. First, this is for one year only, taking into consideration 12 months, not all of which have tornados, or tornados of any length. Second, the correlation, as covered in the last section, is very slim—being .040—which means that there is less than .1 correlation, an extremely low correlation between day and tornado length. If the analyst wants to do a multiple regression, that can be done through this tool, but the columns for the data must be next to each other, so that demonstration will be discussed later. The main purpose for these variables is to ensure that the tools give a consistent result. The main point here is that these two variables are not a good combination for the purpose of regression, given the correlation and the lack of longitudinal data.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.