Data Preparation for Horizon Chart

Data Visualization Advice: Shaping Data for Horizon Chart

To create a horizon chart, you typically need to structure your data in a specific format. Here's a suggested approach along with a sample dataset:

Sample Dataset:

Consider a dataset with columns:

• Date: Time dimension
• Category: Categorical data for grouping
• Value: Numeric data for plotting

Data Shaping Steps:

1. Aggregate Data: If your dataset is at a detailed level, aggregate it to the desired level for the horizon chart.
2. Pivot Data: Reshape the data into a wide format with categories as columns and dates as rows. This can be achieved using tools like pivot_table in Python's pandas library or similar functions in other tools.
3. Normalize Data: Normalize the values within each category to a common scale for effective visualization. This can involve techniques like min-max scaling or z-score normalization.
4. Ensure Data Range: Ensure that your data values fall within a specific range suitable for the horizon chart to display variations clearly.

Code Example (Python - Using pandas):

import pandas as pd

# Sample data
data = {'Date': ['2021-01-01', '2021-01-02', '2021-01-03'],
        'Category': ['A', 'B', 'A'],
        'Value': [100, 150, 120]}

df = pd.DataFrame(data)

# Pivot data
pivot_df = df.pivot_table(index='Date', columns='Category', values='Value').reset_index()

# Normalize data
pivot_df.iloc[:, 1:] = (pivot_df.iloc[:, 1:] - pivot_df.iloc[:, 1:].min()) / (pivot_df.iloc[:, 1:].max() - pivot_df.iloc[:, 1:].min())

print(pivot_df)

By following these steps and shaping your data accordingly, you can prepare it for visualization using a horizon chart effectively.