Report-scatter chart and bubble chart

Scatter chart

Scatter Diter. In regression analysis, data points are distributed on the Cartesian coordinate plane.

A scatter chart shows the general trend in which the dependent variable changes with the independent variable. Based on this, you can select an appropriate function logarithm data point for fitting.

Use two groups of data to form multiple coordinate points, check the coordinate point distribution, and determine whether there is a certain association between the two variables or summarize the coordinate point distribution pattern.

A scatter chart displays the sequence as a group of points. The value is expressed by the position in the chart. The category is represented by different labels in the chart. A scatter chart is usually used to compare aggregate data across classes.

Scatter chart type

• Scatter chart Matrix

It is very troublesome to plot a simple scatter chart for multiple variables at the same time. In this case, the scatter chart matrix can be used to draw a scatter chart between each variable at the same time, so that the main correlation between multiple variables can be quickly found, which is particularly important for multivariate linear regression.

• 3D scatter chart

Although the relationship between multiple variables can be observed in the scatter chart matrix at the same time, some important information may be missed when we observe the scatter chart in two ways. A three-dimensional scatter chart is used to study the relationship between variables in a three-dimensional space determined by three variables, it is often possible to find information that cannot be found in a two-dimensional image.

Notes for scatter chart

Scatter plots are usually used to display and compare values, such as scientific data, statistical data, and engineering data.

Use a scatter chart to compare a large amount of data points without considering the time. The more data a scatter chart contains, the better the comparison effect.

A bubble chart requires that each data point have two values (the top value and the bottom value ).

Scatter plots are ideal for the distribution of processing values and the clustering of data points. If a data set contains many points (for example, thousands of points), a scatter chart is the best chart type. Multiple sequences are displayed in a point chart. In this case, use a line chart instead of a point chart.

By default, a scatter chart displays data points in circles. If there are multiple sequences in a scatter chart, consider changing the markup shape of each point to a square, triangle, diamond, or other shape.

They are scattered in the area from the upper right corner to the lower left corner. We call these two variables positively correlated. There are also some variables, such as the weight of the car and the average distance of every 1l of gasoline consumed by the car, which is negatively correlated. The heavier the car, the shorter the average distance of every 1l of gasoline consumed, the points are scattered in the area from the upper left corner to the lower right corner.

Bubble chart

A bubble chart has the following chart types:

The bubble chart and the three-dimensional bubble chart are similar to the XY scatter chart, but they compare the three values in a group rather than the two values. The third value determines the bubble data point size. You can select a bubble chart or a 3D bubble chart subtype.

 

Report-scatter chart and bubble chart