Segmentation analysis is the process of splitting customers and users into groups based on similar or mutual characteristics.
Homoscedasticity and Heteroscedasticity
Homoscedasticity and heteroscedasticity are important concepts in linear regression
Marketing Attribution Models
Marketing attribution models are ubiquitous in both B2B and B2C spaces, but how do they work?
Statistical Tests For Linear Regression
Statistical tests are essential for validating assumptions in linear regression
Diminishing returns are near-universal in long-term marketing campaigns. This econometric term describes that an output can't continue to increase at the same rate after a certain point.
Seasonality covers everything from seasons to cultural celebrations such as Christmas. Being aware of how business and marketing data is shaped by seasonality unlocks avenues for growth.
Data visualization is fundamental in data science, engineering, analysis, and practically any other skills remit that involves some level of front-end data visibility.
Bias and variance are foundational concepts in machine learning and data science in general. Negotiating the trade-off is essential to building an accurate model.
Linear regression is a simple, dependable technique for analyzing the relationship between linearly related variables.
"Ground truth" is a term borrowed from meteorology. In marketing and data science, the ground truth encapsulates the objective reality behind models, data and predictions.
Kitchen Sink Model
A kitchen sink model includes all variables in a model to see which ones are statistically significant
Bayesian Markov Chain Monte Carlo is a modeling technique that's gaining in popularity as marketing mix modeling
Statistical Significance Calculator
Statistical significance is crucial for evaluating the results of A/B tests
"Garbage in, gargage out" - data scientists and marketers hear that phrase all the time! Cleaning data is the best way to ensure model accuracy and quality.
Search trends provide time and geography-specific data on what people are searching for around the world.
Partial dependence helps explain and visualize the effect of an input variable on a model’s predictions.