Tool based on Matheus Facure Alves Causal Inference for the brave and true handbook. (https://matheusfacure.github.io/python-causality-handbook/13-Difference-in-Differences.html)
This is useful if you can not run an A/B Test
Made by Sterling (https://sterlingdata.webflow.io/)
You will find new examples about how to use Diff-in-Diff.
- Elon Musk Effect
- China's birth control policies
- Aryma Labs - Marketing Mix Modelling
Difference in differences (you’ll find it as DiD, DD, or Diff-in-Diff as well) is a statistical technique used in econometrics and quantitative research. This will be done as a natural experiment when there’s no possible any randomnization. It calculates the effect of a treatment (i.e., an explanatory variable or an independent as more sunlight exposure to sunflower ) on an outcome (i.e., a response variable or dependent variable as number of seeds) by comparing the average change over time in the outcome variable for the treatment group to the average change over time for the control group. This method may still be subject to certain biases (e.g., mean regression, reverse causality and omitted variable bias).
Here's you'll find a streamlit app code and an example data file to learn how to use it.
The SterlingDiffInDiffTool is a web-based application designed to facilitate Difference-in-Differences (DiD) analysis, a statistical method used to estimate causal relationships in observational studies. Leveraging Streamlit, the tool provides an interactive interface for users to input data, perform DiD analysis, and visualize results without the need for extensive programming knowledge. This makes causal inference techniques more accessible to researchers and analysts.
#How the project came about
This project draws inspiration from Matheus Facure Alves's "Causal Inference for the Brave and True" handbook, aiming to translate theoretical concepts into a practical tool. Developed by Sterling Data, the tool addresses the need for accessible applications that allow users to conduct DiD analysis without extensive coding.
#Motivation
In many real-world scenarios, conducting randomized controlled trials is impractical or unethical. The DiD method offers a viable alternative for estimating treatment effects using observational data. However, implementing DiD analyses can be challenging for those without a programming background. This tool seeks to bridge that gap by providing an easy-to-use platform for conducting DiD analyses.
#Limitations
Users can upload their datasets into the application, specify treatment and control groups, and define the time periods for analysis. The tool then performs the DiD analysis and presents the results through tables and visualizations, aiding in the interpretation of the treatment effect.
#Challenges
Users can upload their datasets into the application, specify treatment and control groups, and define the time periods for analysis. The tool then performs the DiD analysis and presents the results through tables and visualizations, aiding in the interpretation of the treatment effect.
#What problem it hopes to solve
Users can upload their datasets into the application, specify treatment and control groups, and define the time periods for analysis. The tool then performs the DiD analysis and presents the results through tables and visualizations, aiding in the interpretation of the treatment effect.
#What the intended use is
Users can upload their datasets into the application, specify treatment and control groups, and define the time periods for analysis. The tool then performs the DiD analysis and presents the results through tables and visualizations, aiding in the interpretation of the treatment effect.