Using a mix of computer programming and database design, Maderas Partners excels at taming messy data and performing complex calculations on datasets of any size. We rely heavily on R and SQL and also incorporate other languages, such as Python and Julia, as necessary.
Sometimes, the biggest challenge is creating a uniform dataset. Here we outline one recent example.
From program evaluation to time series analysis to predictive modeling, Maderas Partners got its start in statistical and econometric analysis. We perform the majority of our econometric work in Stata and machine learning in Python.
The Difference-in-Difference (DiD) analysis is a powerful econometric tool to analyze the difference between two groups after some type of treatment.
Using actual data from the Seattle Police Dept, we created a machine learning model to predict the likelihood of arrest after routine police stops.
We understand that while fancy data analytics is great, sometimes what the client really needs is to communicate the results in a way that is easy to understand. Employing Power BI and Tableau, we can transform a collection of raw numbers into beautiful charts and graphs that allow for robust storytelling.
The Machine Learning portion of the K-Nearest Neighbors exercise.
Data Analytics & Investment