GWR

Introduction

We often use regression technologies to analyze the dataset to explore the potential relationship between different factors. Regression analysis can not only explore and construct the model based on the spatial relationship, but also can explain the observed features hiding behind the spatial model; therefore, users can explicitly understand why some events happening on a specific location and easily predict the uncertain situation based on these potential effects.

‌Currently, our platform supports two regression models, Linear Regression and Geographically Weighted Regression. Linear Regression is a common regression method providing a global model to analyze the relationship between dependent variables and explanatory variables over an entire space. However, geographic spatial features highly depend on the spatial autocorrelations; therefore, the result generated by traditional regression methods become highly unreliable. In order to solve the problem, Geographically Weighted Regression construct a stable local regression function for each spatial feature inside the dataset.

1. Select the regression type

• Regression Type: Linear Regression or Geographically Weighted Regression.

2. Set the input data

• Input Data: Spatial Data only.

3. Select the dependent variable

• Dependent Variable: Its values are dependent on the values of the explanatory variables.

4. Select the explanatory variables

• Explanatory Variables: The variables could affect the values of the dependent variable.

5. Set the output data

• Output Data Name: If no name is set, the default is “toolName_time”.

6. Submit

After completing the above settings, click Submit to start the tool.
When it finishes successfully, a message will be shown at the top of the page. You can click the Open Data button to start accessing the data or go to the Data page to view it.