Skip to content

Visualization API Reference

FracTime provides interactive visualizations using wrchart.

Overview

import fractime as ft

# Plot any FracTime object
ft.plot(result)
ft.plot(analyzer)
ft.plot(analyzer.hurst)

# Plot forecast with Monte Carlo paths
ft.plot_forecast(prices, result)

plot()

The universal plotting function that handles any FracTime object.

ft.plot(
    obj,            # Object to plot
    view=None,      # For Metric: 'point', 'rolling', 'distribution'
    title=None,     # Custom title
    show=True,      # Display immediately
    **kwargs        # Additional wrchart arguments
)

Parameters

Parameter Type Default Description
obj various required FracTime object to plot
view str None View type for Metric objects
title str None Custom plot title
show bool True Display immediately
width int varies Chart width in pixels
height int varies Chart height in pixels
theme str 'dark' Theme ('dark' or 'light')

Supported Objects

Object Type Default Visualization
ForecastResult Forecast with confidence bands
AnalysisResult Analysis dashboard
Analyzer Analysis dashboard
Metric Auto-detect best view

Returns

wrchart chart object (ForecastChart, MultiPanelChart, or Chart).

plot_forecast()

Plot forecast with Monte Carlo paths and path density visualization.

ft.plot_forecast(
    prices,                 # Historical price data
    result,                 # Forecast result dict
    dates=None,             # Historical dates (optional)
    title=None,             # Chart title
    colorscale='viridis',   # Color scale for path density
    show_percentiles=True,  # Show percentile lines
    **kwargs                # Additional arguments
)

Parameters

Parameter Type Default Description
prices np.ndarray required Historical price data
result dict required Forecast result with paths, probabilities
dates np.ndarray None Historical dates
title str None Chart title
colorscale str 'viridis' Color scale ('viridis', 'plasma', 'inferno', 'hot')
show_percentiles bool True Show percentile lines

Returns

ForecastChart object.

Plotting ForecastResult

model = ft.Forecaster(prices)
result = model.predict(steps=30)

# Basic plot
ft.plot(result)

# Custom title
ft.plot(result, title="30-Day Price Forecast")

# Don't show immediately
chart = ft.plot(result, show=False)
chart.to_html("forecast.html")

What's Shown

  • 95% confidence band (outer band)
  • 50% confidence band (inner band)
  • Median forecast (solid line)
  • Mean forecast (dashed line)

Plotting AnalysisResult

analyzer = ft.Analyzer(prices)

# Plot analysis dashboard
ft.plot(analyzer)

# Or use the result directly
ft.plot(analyzer.result)

What's Shown

  • Hurst gauge - Value from 0 to 1
  • Fractal dimension gauge - Value from 1 to 2
  • Volatility gauge - Annualized percentage
  • Regime bar chart - Probability distribution

Plotting Metric

Metrics support three views:

Auto-Detect

ft.plot(analyzer.hurst)  # Auto-selects best view

Selection logic: 1. If rolling data exists → rolling view 2. Else if distribution exists → distribution view 3. Else → point view

Point View

Gauge chart showing the point estimate.

ft.plot(analyzer.hurst, view='point')

Rolling View

Time series of rolling values.

ft.plot(analyzer.hurst, view='rolling')

Shows: - Rolling values over time - Line chart with the metric values

Distribution View

Histogram of bootstrap samples.

ft.plot(analyzer.hurst, view='distribution')

Shows: - Bootstrap distribution as bar chart - Point estimate in title

Customization

wrchart Arguments

Pass additional arguments to wrchart:

ft.plot(
    result,
    title="My Forecast",
    height=600,
    width=1000,
    theme='dark',  # or 'light'
)

Common Layout Options

Option Description
height Figure height in pixels
width Figure width in pixels
theme Theme ('dark' or 'light')
colorscale For forecasts: 'viridis', 'plasma', 'inferno', 'hot'

Saving Charts

HTML (Interactive)

chart = ft.plot(result, show=False)
chart.to_html("forecast.html")

Examples

Forecast Plot

import fractime as ft

model = ft.Forecaster(prices)
result = model.predict(steps=30)

ft.plot(result, title="30-Day Forecast")

Forecast with Monte Carlo Paths

import fractime as ft

forecaster = ft.FractalForecaster(lookback=252)
forecaster.fit(prices)
result = forecaster.predict(n_steps=30, n_paths=500)

chart = ft.plot_forecast(
    prices[-100:],
    result,
    title="Monte Carlo Forecast",
    colorscale='plasma'
)
chart.to_html("forecast.html")

Analysis Dashboard

analyzer = ft.Analyzer(prices)
ft.plot(analyzer, title="Fractal Analysis")

Rolling Hurst

analyzer = ft.Analyzer(prices, dates=dates, window=63)
ft.plot(analyzer.hurst, view='rolling', title="Rolling Hurst Exponent")

Bootstrap Distribution

analyzer = ft.Analyzer(prices, n_samples=2000)
ft.plot(analyzer.hurst, view='distribution', title="Hurst Distribution")

Save to HTML

chart = ft.plot(result, show=False)
chart.to_html("forecast.html")

Custom Styling

ft.plot(
    result,
    title="Price Forecast",
    height=500,
    width=900,
    theme='light',
)

Working with Charts

The returned chart is a wrchart chart object:

chart = ft.plot(result, show=False)

# Display the chart
chart.show()

# Save to HTML
chart.to_html("forecast.html")

# In Streamlit
chart.streamlit()

For ForecastChart, you can chain configuration:

chart = ft.plot_forecast(prices, result, show=False)
chart.colorscale('plasma')
chart.show_percentiles(True)
chart.show_weighted_forecast(True)
chart.show()

See Also