Robyn
EconML
Robyn | EconML | |
---|---|---|
14 | 8 | |
1,035 | 3,550 | |
1.8% | 1.3% | |
9.1 | 8.5 | |
7 days ago | 8 days ago | |
Jupyter Notebook | Jupyter Notebook | |
MIT License | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Robyn
-
Mixed Marketing Modeling Approach for attribution?
With all this talk about Google and other platforms deprecating 3P tracking in favor of more aggregate "tracking", my team is considering a marketing mix modeling tool. One that comes to mind is this tool - Robyn
-
Is Hierarchical Bayesian Modelling used in industry?
Robyn - R Library
-
Can Ads Be GDPR Compliant?
Furthermore, both Google and Meta have quietly conceded that a lot of the digital attribution data they generate is pretty bunk. It’s why Meta developed Robyn, which uses MMM techniques that have long been used to measure effectiveness of traditional channels: https://github.com/facebookexperimental/Robyn
-
Thoughts on Exposure Metrics Based Marketing Mix Modeling (MMM)?
Here's my first question - are there other approaches that model media spending and media exposure metrics separately? I have noticed that Robyn has included both exposure metrics and media spending and got them transformed via a non-linear model called the Michaelis-Menten function to establish the spend-exposure relationship, as you can find it here. What if we want, or is it possible to keep only media exposure metrics instead of media spending? If yes, what would be the alternative approaches apart from Logarithmic Regression, Ridge Regression, Bayesian approach, etc?
-
My Favorite Off-the-Shelf Data Science Repos, What Are Yours?
Here are my top off-the-shelf data science models for Marketing. Would be interested which other marketing data science tools you use?
Product Recommendation on Your Website with Metarank (https://github.com/metarank/metarank)
Metarank is a tool that helps you easily build an advanced recommendation engine for your products or content on your website. To get started you only need historical performance data of your products (e.g. number of clicks) and additional metadata like product rating, genre, ingredients or price. In a YAML file, you define the features and the model parameters (e.g. number of iterations, modeling technique). The API service integrates with Apache Flink and can be easily integrated into Kubernetes clusters.
User Journey Analysis on your Website with Retentioneering (https://github.com/retentioneering/retentioneering-tools)
Retentioneering helps you to understand the user journey on your website. Retentioneering is a Python library that allows you to easily connect your Google Analytics data (in Bigquery). You define user-id, event-type and time stamp. From this data input a comprehensive graph network is created with gains and losses as you know it from a customer journey. In addition, customer segments are created that have a similar customer journey. This reduces the complexity of a purely descriptive view of the data.
Marketing Mix Modeling with Robyn (https://github.com/facebookexperimental/Robyn)
Less third-party cookie means less attribution models. The answer to this is Marketing Mix Modeling. Marketing mix models are regression models that use statistical probability to calculate the effect size of marketing channels and other independent variables. The advantage is that business context can be modeled much more realistically. For example, Google Searches for the own brand can be integrated to determine the share of the own brand strength in the revenue. Likewise, offline advertising measures can be modeled with other metrics in this context (e.g. offline advertising with GRPs). Robyn takes into account adstock effects, ROAS calculation and multicollinarity in the marketing channels. In addition, with simple functionality, budgets can be optimized using the predictions and results from marketing tests can be integrated into the model for calibration.
-
Marketing Spend Optimization
Try this https://facebookexperimental.github.io/Robyn/ And for more advanced project you can try this https://siyasgupte.medium.com/causal-impact-understand-the-inner-workings-to-optimize-your-results-506ed442619
-
data engineering should not be an issue for Data Scientists
For Data Scientists super interesting but still in WiP: We are implementing a advanced Marketing Mix Model soon. It combines bascially Prophet + Ridge Regression + Nevergrad (it is already a R library from Meta: https://github.com/facebookexperimental/Robyn
- Reviewing Marketing Mix Model - concerned about a few issues...
-
Media Mix Modeling. What variables should I use? What would be a good R^2?
Take a look at Robyn by Facebook Meta, if you haven't
-
Data Science and Marketing
MMM (R): Robyn
EconML
-
[D] What approach to decide which class is most optimal for recovery?
A good package with many of the tools used for this type of problem as well as pretty good documentation about how it all works is https://github.com/microsoft/EconML
- Getting treatment effects from a random forest
-
Data Science and Marketing
Uplift Modeling (python): CausalML, EconML
-
EconML: A Python Package for ML-Based Heterogeneous Treatment Effects Estimation
Github: https://github.com/microsoft/EconML
-
UpliftML: An uplift modeling library that handles web scale datasets
Many libraries have recently emerged that offer implementations of algorithms for heterogeneous treatment effect estimation (or, CATE estimation). The most well-known examples are Microsoft's EconML (https://github.com/microsoft/EconML) and Uber's CausalML (https://github.com/uber/causalml). Existing libraries require all data to fit in memory, which is often a limitation for industry applications on web scale datasets. Booking.com's new library offers similar functionality on top of Spark, enabling web scale uplift modeling.
-
[q] before/after test
EconML and CausalLift are pretty good python packages that help you build uplift models. scikit-uplift is a decent sklearn style wrapper package that can be helpful as well. One of the drawbacks of these packages is they only allow for the modeling of a single treatment. mr-uplift is a newer package that allows you to model the multiple treatment effects simultaneously. I haven't used it personally, but it does look fairly interesting.
-
[N] Spotify Confidence - open source for analyzing a/b test data
Can't see how this adds to decades of causal inference packages development in stats oriented frameworks like R/Stata/EViews etc and the ongoing effort of porting this to Python. If you want something fancy there's already EconML.
-
What are some applications of Data Science in Digital Marketing?
Uplift Modeling - This is a very powerful technique aimed at discovering the customers who are most likely to respond to your marketing efforts. Some good python libraries for this are EconML and mr-uplift
What are some alternatives?
Prophet - Tool for producing high quality forecasts for time series data that has multiple seasonality with linear or non-linear growth.
causalml - Uplift modeling and causal inference with machine learning algorithms
MMM-BurnIn
upliftml - UpliftML: A Python Package for Scalable Uplift Modeling
causalglm - Interpretable and model-robust causal inference for heterogeneous treatment effects using generalized linear working models with targeted machine-learning
lightweight_mmm - LightweightMMM 🦇 is a lightweight Bayesian Marketing Mix Modeling (MMM) library that allows users to easily train MMMs and obtain channel attribution information.
causallift - CausalLift: Python package for causality-based Uplift Modeling in real-world business
mmm_stan - Python/STAN Implementation of Multiplicative Marketing Mix Model, with deep dive into Adstock (carry-over effect), ROAS, and mROAS
tensor-house - A collection of reference Jupyter notebooks and demo AI/ML applications for enterprise use cases: marketing, pricing, supply chain, smart manufacturing, and more.
mta - Multi-Touch Attribution
Robyn - Robyn is a Super Fast Async Python Web Framework with a Rust runtime.