Language

Creating Graphs

Create graphs to visualize your data. This page explains basic graph types including histograms, scatter plots, bar charts, and time series plots.

How to Create Graphs

  1. Open Graph Builder tab: Select Analysis → Graph Builder... from the menu bar
  2. Select dataset: Choose the dataset to graph from the dropdown menu in the header
  3. Select graph type: Click the type of graph you want to display (Histogram, Scatter Plot, Bar Chart, etc.)
  4. Configure settings: Set axes and variables according to the selected graph type
    • Histogram: Variable, number of bins, grouping
    • Scatter plot: X-axis, Y-axis, color coding
    • Bar chart: Category, value, aggregation method, sorting
  5. Adjust options: Set filter conditions and additional options as needed

Graphs are displayed in real-time in the preview area on the right, allowing you to explore optimal visualizations while changing settings.

Basic Graphs

Histogram

Displays the distribution of a single numeric variable as a bar graph. Useful for understanding frequency distribution and shape of data, and identifying outliers.

Use cases:

  • Check age distribution
  • Examine test score distribution
  • Understand sales amount distribution

Settings:

Basic tab:

  • Column: Select the numeric column to graph
  • Bins: Specify the number of bars (bins)
  • Orientation: Vertical or horizontal bars
  • Show Density Curve: Overlay a density curve on the histogram

Grouping tab:

  • Group By: Select a category column to display color-coded by group

Annotations tab:

  • Show Mean: Display the mean as a solid line
  • Show Median: Display the median as a dashed line

Histogram of numeric variable distribution created with MIDAS: displays data frequency per bin and visualizes distribution shape and outliers

Scatter Plot

Visualizes the relationship between two numeric variables as points. Discover correlations, patterns, clusters, and outliers. Also supports density visualization and grouping by categorical variables.

Use cases:

  • Examine relationship between height and weight
  • Check correlation between advertising spend and sales
  • View relationship between scores in two exam subjects

Settings:

Basic tab:

  • X-Axis: Numeric column for horizontal axis
  • Y-Axis: Numeric column for vertical axis
  • Color: Select a category column for color-coded grouping

Density tab:

  • Display Mode: Select display mode
    • Points only: Show points only (for small datasets)
    • Density only: Show density only
    • Points + Density (Outliers Only): Show density with outlier points
    • Points + Density (All Points): Show density with all points

Reference Lines tab:

  • Add Reference Line: Adds reference lines to mark specific values or thresholds

MIDAS Scatter Plot: places two numeric variables on X and Y axes, displaying each data point as a dot. Used for discovering correlations and patterns

Bar Chart

Compares values by category using vertical or horizontal bars. Ideal for comparing categorical data, ranking displays, and showing aggregate statistics (count, sum, average, etc.).

Use cases:

  • Compare sales by product category
  • Display customer count by region
  • Aggregate order count by month

Settings:

  • Category Column: Column representing categories
  • Values: Select a numeric column and an aggregation method (Sum, Average, Min, or Max). Choose Count (Frequency) instead to count rows per category
  • Orientation: Vertical or horizontal bars
  • Sort Order: Sort order (alphabetical, value descending/ascending)
  • Top N Categories: Display only the top N categories
  • Show value labels: Display values on top of bars

MIDAS Bar Chart: displays values by category as vertical or horizontal bars, with selectable aggregation methods (count, sum, average, etc.)

Time Series Plot

Displays data changes over time as a line chart. Identify trends, seasonality, and patterns. Display multiple time series with different line styles and compare across periods.

Use cases:

  • Check daily sales trends
  • Track monthly visitor changes
  • Observe annual temperature variations

Settings:

Basic tab:

  • X-Axis (Time): Column representing date or datetime
  • Y-Axis (Values): Numeric columns to display (multiple selection supported)

Ranges tab:

  • Custom Range Bands: Add bands to highlight specific periods or ranges

Advanced tab:

  • Chart Title: Set the graph title

MIDAS Time Series Plot: displays data changes along time axis as line chart, visualizing trends and seasonal patterns

Other Graphs

Pair Plot

Displays pairwise relationships between multiple numeric variables as a scatter plot matrix. The diagonal shows histograms, and off-diagonal cells show scatter plots with correlation coefficients.

Use cases:

  • Check correlations between multiple measurements at once
  • Understand the overall structure of a dataset

Settings:

  • Variables: Select numeric columns via checkboxes (interval/ratio scale columns, minimum 2 variables). A warning is shown when selecting more than 6 variables due to performance and readability impact
  • Display Order: Change the display order of selected variables

MIDAS Pair Plot: displays pairwise relationships between multiple numeric variables as scatter plot matrix. Shows distribution of each variable on diagonal, allowing simultaneous confirmation of multivariate correlations

Datetime Histogram

Visualizes the distribution of datetime data as a histogram on a time axis. Aggregates event frequency by time interval to identify temporal patterns.

Use cases:

  • Check log event frequency by time period
  • Visualize concentration of access times
  • Analyze temporal distribution of orders and transactions

Settings:

  • Date/Time Column: Column representing datetime or date
  • Time Interval: Aggregation time interval (Auto, 1 minute to 1 year)
  • Show trend line: Display a trend line using LOESS smoothing

MIDAS Datetime Histogram: displays distribution of datetime data as histogram on time axis, visualizing event frequency and temporal patterns

Custom Graph

Creates flexible visualizations based on Grammar of Graphics principles. Build layered plots by combining geometric elements, statistical transformations, and aesthetic mappings.

See the Custom Graph Guide for details.

Graph Operations

Interactive Selection

All graphs support data point selection. Selection methods vary by graph type:

Rectangle Selection (Scatter Plot, Histogram, Datetime Histogram)

  • Select mode: Drag mouse to select a range
  • Pan mode: Drag graph to move the display range

Mode toggle buttons are displayed in the bottom-right corner of the graph.

Click Selection (Bar Chart, Time Series Plot, Pair Plot, Custom Graph)

  • Click bars, points, or bins to select
  • Ctrl/Cmd+click to add to selection or deselect

Selected rows are highlighted in other tabs such as Data Table, Statistics, and Selected Rows.

Saving Graphs

Graph settings created in Graph Builder are automatically saved to the project as tab state. Graph settings are preserved when you close and reopen the project.

Changing Settings

Graph settings (column selection, filters, display options, etc.) can be changed at any time. Graphs update in real-time when settings are changed.

Next steps

  • Custom Graph - Flexible visualization with Grammar of Graphics
  • Report - Compile graphs and tables into reports

See also