Lee Richardson
YOU?
Author Swipe
View article: Choosing a Proxy Metric from Past Experiments
Choosing a Proxy Metric from Past Experiments Open
In many randomized experiments, the treatment effect of the long-term metric (i.e. the primary outcome of interest) is often difficult or infeasible to measure. Such long-term metrics are often slow to react to changes and sufficiently noi…
View article: O10 3D-designed custom-made modular headgear for Children using non-invasive ventilation. [The ‘COMFORT’ project: Custom-Made Facemasks for Respiratory Therapy]
O10 3D-designed custom-made modular headgear for Children using non-invasive ventilation. [The ‘COMFORT’ project: Custom-Made Facemasks for Respiratory Therapy] Open
Non-invasive ventilation (NIV) is assisted respiratory support delivered via facemask for people with chronic respiratory failure. Commercial NIV masks are available but masks that fit well are difficult to find for children who have smal…
View article: Choosing a Proxy Metric from Past Experiments
Choosing a Proxy Metric from Past Experiments Open
In many randomized experiments, the treatment effect of the long-term metric (i.e. the primary outcome of interest) is often difficult or infeasible to measure. Such long-term metrics are often slow to react to changes and sufficiently noi…
View article: Pareto optimal proxy metrics
Pareto optimal proxy metrics Open
North star metrics and online experimentation play a central role in how technology companies improve their products. In many practical settings, however, evaluating experiments based on the north star metric directly can be difficult. The…
View article: Surrogate for Long-Term User Experience in Recommender Systems
Surrogate for Long-Term User Experience in Recommender Systems Open
Over the years we have seen recommender systems shifting focus from optimizing short-term engagement toward improving long-term user experience on the platforms. While defining good long-term user experience is still an active research are…
View article: Going deep: models for continuous-time within-play valuation of game outcomes in American football with tracking data
Going deep: models for continuous-time within-play valuation of game outcomes in American football with tracking data Open
Continuous-time assessments of game outcomes in sports have become increasingly common in the last decade. In American football, only discrete-time estimates of play value were possible, since the most advanced public football datasets wer…
View article: Continuously Updated Data Analysis Systems
Continuously Updated Data Analysis Systems Open
When doing data science, it's important to know what you're building. This paper describes an idealized final product of a data science project, called a Continuously Updated Data-Analysis System (CUDAS). The CUDAS concept synthesizes idea…
View article: The Sliding Window Discrete Fourier Transform
The Sliding Window Discrete Fourier Transform Open
The discrete Fourier transform (DFT) is a widely used tool across science and engineering. Nevertheless, the DFT assumes that the frequency characteristics of a signal remain constant over time, and is unable to detect local changes. Resea…
View article: Augmenting Adjusted Plus-Minus in Soccer with FIFA Ratings
Augmenting Adjusted Plus-Minus in Soccer with FIFA Ratings Open
In basketball and hockey, state-of-the-art player value statistics are often variants of Adjusted Plus-Minus (APM). But APM hasn't had the same impact in soccer, since soccer games are low scoring with a low number of substitutions. In soc…
View article: The Sliding Window Discrete Fourier Transform
The Sliding Window Discrete Fourier Transform Open
This paper introduces a new tool for time-series analysis: the Sliding Window Discrete Fourier Transform (SWDFT). The SWDFT is especially useful for time-series with local- in-time periodic components. We define a 5-parameter model for noi…
View article: An Algorithm for the 2D Radix-2 Sliding Window Fourier Transform.
An Algorithm for the 2D Radix-2 Sliding Window Fourier Transform. Open
We present a new algorithm for the 2D Radix-2 Sliding Window Fourier Transform (SWFT). Our algorithm avoids repeating calculations in overlapping windows by using a tree representation of the Cooley-Tukey Fast Fourier Transform (FFT). For …
View article: Indonesia Synthetic Ecosystem
Indonesia Synthetic Ecosystem Open
Indonesia synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: South Dakota Synthetic Ecosystem
South Dakota Synthetic Ecosystem Open
South Dakota synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; …
View article: Kenya Synthetic Ecosystem
Kenya Synthetic Ecosystem Open
Kenya synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: Iowa Synthetic Ecosystem
Iowa Synthetic Ecosystem Open
Iowa synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 2013 and…
View article: Fiji Synthetic Ecosystem
Fiji Synthetic Ecosystem Open
Fiji synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: Arkansas Synthetic Ecosystem
Arkansas Synthetic Ecosystem Open
Arkansas synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 2013…
View article: Philippines Synthetic Ecosystem
Philippines Synthetic Ecosystem Open
Philippines synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: Tennessee Synthetic Ecosystem
Tennessee Synthetic Ecosystem Open
Tennessee synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 201…
View article: New Mexico Synthetic Ecosystem
New Mexico Synthetic Ecosystem Open
New Mexico synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 20…
View article: Indiana Synthetic Ecosystem
Indiana Synthetic Ecosystem Open
Indiana synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 2013 …
View article: Pakistan Synthetic Ecosystem
Pakistan Synthetic Ecosystem Open
Pakistan synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: Connecticut Synthetic Ecosystem
Connecticut Synthetic Ecosystem Open
Connecticut synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 2…
View article: Nevada Synthetic Ecosystem
Nevada Synthetic Ecosystem Open
Nevada synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 2013 a…
View article: Cambodia Synthetic Ecosystem
Cambodia Synthetic Ecosystem Open
Cambodia synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: North Carolina Synthetic Ecosystem
North Carolina Synthetic Ecosystem Open
North Carolina synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data…
View article: Jordan Synthetic Ecosystem
Jordan Synthetic Ecosystem Open
Jordan synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: Nigeria Synthetic Ecosystem
Nigeria Synthetic Ecosystem Open
Nigeria synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: Iraq Synthetic Ecosystem
Iraq Synthetic Ecosystem Open
Iraq synthetic ecosystem dataset consisting of tables for persons and households. This dataset was created by the SPEW R Package using the following input data sources: GeoHive counts and IPUMS-I shapefiles and microdata.
View article: New York Synthetic Ecosystem
New York Synthetic Ecosystem Open
New York synthetic population dataset consisting of tables for persons, households, schools, and workplaces. This dataset was created by the SPEW program from 2010 ACS SF 5-Year counts; 2010 US TIGER Roads; ACS PUMS 1-Year micro data; 2013…