Causalify Logo
Causalify (Proof of Concept)

Marketing Case Study with Synthetic Control

Example application of the synthetic control method in the tech industryReplication by José Alvarez

This interactive page replicates key results from the hypothetical marketing case study in Causal Inference in Python (Chapter 9: Synthetic Control). Utilizing the synthetic dataset meticulously crafted by Facure (2023) for this illustrative purpose, users can explore the effects of marketing strategies on app download trends in São Paulo.

The Synthetic Control Method (SCM) is a statistical technique to estimate the causal effect of a treatment, policy, or intervention on a particular unit (such as a country, region, or firm) by constructing a synthetic control group. This method is particularly useful if there is one (or just a few) treated units, and if it's not possible to rely on experimental designs.

Click on "Examples" (at the top of the page) to replicate results from other published works. Users can also upload their own datasets for custom analysis and generate insightful visual comparisons.

Choose CSV

Unit Variable

Identify units and controls

Time Variable

Select time variable

Predictors

Select...

Dependent Variable

Dependent variable

Special Parameters (Optional)

Select Covariate
Select Time
Select Operation

0%

0%


Current Features

  • Synthetic Control Plot
  • Synthetic Control Weights Table
  • Gap Plot (Treated - Synthetic)
  • Post-treatment Impact Table
  • Predictor Means Table

Upcoming Features

  • Placebo-in-time capability
  • Placebo-in-space capability
  • P-values
  • Customizable PDF Generation
  • Link-based Report Sharing
  • Parallel Computation and Faster CPUs
  • Causal Inference Expert AI Chatbot
  • AI-powered Output Analysis & Interpretation
  • World Bank API Integration
  • Major Country/State Level Data Sources Integration
This is a prototype and proof of concept.
We value your feedback and suggestions.
contact@causalify.com

Accelerate your causal inference workflow

In our early stages, we value your input. Cast your vote for your preferred feature or suggest a new one.

Causalify enables impact evaluation for enterprise

Leverage synthetic controls with Causalify's no-code interactive web app.
Synthetic Control Plot
Synthetic Control Plot

Evaluate the impact of an intervention

Synthetic Control Weights
Synthetic Control Weights

Understand the contribution of each control unit

Gap Plot (Treated - Synthetic)
Gap Plot (Treated - Synthetic)

Visualize the size of the causal effect

Post-treatment Impact Table
Post-treatment Impact Table

Compare average vs. aggregated impacts

Predictor Means Table
Predictor Means Table

Analyze the pre-treatment covariates

P-value
P-value

Find the statistical significance of the results

Upcoming features

In-time Placebos

Test the validity of your results relative to time

In-space Placebos

Test the validity of your results among all controls

PDF Report Generation

Customize and export your results

Link-based Report Sharing

Share an interactive presentation with a link

Parallel Faster Computation

Analyze your data faster and iterate efficiently

Causal Inference AI Expert

Meet CausalGPT: Causalify's AI assistant

AI-powered Output Analysis

Understand and interpret your results better

World Bank API Integration

Access the World Bank's global data with ease

Data Sources Integration

Access major country / state level sources

Follow Causalify

Keep posted on our latest updates and tutorials.