Statistical
Analysis (print
topic)
Grunt’s Statistical Analysis module lets you run simple or multiple linear regression analyses quickly using your own verified data. As soon as the app has enough data to generate the Statistical Analysis, the app switches to the Analysis tab so you can see the results. The results update immediately as you change your variables, allowing you to try different combinations without switching between screens, programs, or forms.
The form is made up of a control column on the left side and a row of tabs across the top. At the bottom of the form, a status bar will show a hint based on which control you have selected.
The tabs across the top allow you to select different panels for Input or Output.
- Simple linear regression analyses are generated when only one field is checked in the Fields to analyze selections.
- Multiple linear regression analyses are generated when more than one field has been selected.
Dependent Variable Three optional units of comparison are
presented for you to analyze: Sale Price, Price/SF GLA,
and Price/Ac land. Select the u
nits
of value you want your result in. The default setting is Sale
Price.
Confidence Level [of the coefficients calculated] is always
generated for 95%. If you want to calculate a different confidence
level, check the box and
enter the percentage confidence level sought in the edit box.
Fields selector identifies which independent variable(s)
to analyze statistically.
Example: Develop the equation that relates the price of a property given the size of its site, the quality of the improvements, and the number of bedrooms. Your dependent variable option would be Sale Price, and you would check the boxes in the Fields SITE, QUAL, and BR. The Analysis tab will show, and you’ll see your results.
Tabbed Sheets
The grid shown in this tab lets you see your input data. Some records may appear in the list that are simply not suitable for the analysis, although they would meet the general tests for inclusion in the dataset. Your dataset could be skewed if it included properties that are outliers for such things as property specific blight or enhancement (George Washington slept here, site of a violent crime, etc). To omit specific records from your analysis, highlighting them and click the delete or backspace key. This will place an asterisk (*) next to their CompNumber indicating that they are ignored for the analysis. To unmark (include) them, select them and click the enter or return key.
The table shows only the columns to used for the analysis. The image to the right shows what it may look like given our example for the Fields selector and with the sales omitted as shown in the Source Data Tab section.
The table in the Analysis Tab reports the results of your analysis. It follows the same format as the analysis you would get from Microsoft’s popular Excel® spreadsheet program.
Right-clicking the table lets you save the report in CSV format or copy it to the clipboard for pasting directly into other programs. You can also altering your inputs if you’re not satisfied with the results, and the table will update automatically.
Contains handy general references for the numbers contained in the Summary Output shown in the Analysis Tab grid.