Search⌘ K
Home
Courses
Mastering Plotly and Dash Data Visualization with Python

Visualizing Regression Plots

Explore how to visualize regression models and smoothing techniques with Plotly. Understand scatter plot smoothing, linear regression, LOWESS curves, moving averages, exponential smoothing, and 3D scatter plots to analyze complex datasets effectively.

We'll cover the following...

Scatter plot smoothing

Scatter plot smoothing is a statistical technique used to gain an understanding of the general patterns in the data by creating a curve with a smoothed estimate of the relationship between two variables, x and y. It is also particularly useful for assessing bivariate outliers and influential observations that may affect the relationship between the variables.

This is a macroeconomic datasetJ. W. Longley (1967) An appraisal of least-squares programs from the point of view of the user. Journal of the American Statistical Association 62, 819–841. recording economic variables of:

  • GNP.deflator: GNP implicit price deflator (1954=100)

  • GNP: Gross National Product

  • Unemployed: Number of unemployed

  • Armed.Forces: Number of people in the armed forces

  • Population: ‘noninstitutionalized’ population ≥ 14 years of age

  • Year: The year (time)

  • Employed: Number of people employed

Note: This dataset is often used to demonstrate how regressing Employed on the remaining variables causes multicollinearity.

Python 3.8
# Import libraries
import pandas as pd
import numpy as np
# Import datasets
longley = pd.read_csv('/usr/local/csvfiles/longley.csv')
print(longley.head())

Linear regression

Plotly Express allows us to fit a linear regression to a dataset with an x and y variable. We do this by placing a trendline="ols" argument to specify that we wish to use an ordinary least squares regression model. We can set the color of the line using trendline_color_override. By hovering over the trend line, we gain an insight into the slope, intercept, coefficient of determination (R2R^2), and prediction y^\hat{y} ...