Posts

Data science discussions and general updates from Conlan Scientific.

Use Machine Learning To Grow Your Lending Business

This post will quickly outline how your can financing business can use machine learning to ... Document and execute a consistent lending strategy. Discover which attributes most significantly affect deal performance. Reduce your susceptibility to fraud from both internal and external actors. What Are We Doing? …

Yahoo Finance Recognizes Our Achievements in Financial AI

Link to Yahoo Finance article: https://finance.yahoo.com/news/conlan-scientific-named-top-ai-224600809.html

D.C. is the Data Science Capital of the U.S.

... at least according to Google Search Trends it is. Google places Washington D.C. first in the country for interest in data science out of all 50 U.S. states. We can only speculate what gave us this score. Is it our thriving EdTech industry? Is it biostatistics at …

Fundamental Analysis of Real Estate Stocks

This scatterplot uses t-SNE analysis to visualize the relationships between financial performance and market pricing of public real estate companies. Select from the industries within the real estate sector to explore more relationships. While t-SNE is not technically a clustering algorithm, it can help us visualize …

Conlan Scientific Partners with UVA for Data Science Course

Conlan Scientific is proud to announce that it has partnered with the University of Virginia Department of Statistics to teach STAT 4559: Data Science Practicum during the Spring 2018 semester. The Arrangement Chris Conlan and Professor Jeff Holt will co-teach the class to a small group of students. Conlan Scientific …

Cybersecurity Stock Portfolio

This plot simulates a theoretical constant-weighted portfolio of cybersecurity stocks over the past five years. Try adjusting both the performance metrics and the asset allocations to see how the portfolio could have performed in the past. With constant weights, it is difficult to push the Sharpe …

Avoiding Data Leakage in Machine Learning

To properly evaluate a machine learning model, the available data must be split into training and test subsets. Data leakage occurs when, in one way or another, information regarding the test set inappropriately influences the training or evaluation of the model. This causes us to overestimated the performance of a …

Misconceptions about P-values

P-values are currently the most common metric used to draw conclusions from research. Given this, one would expect they are well-understood and non-controversial. Nothing could be further from the truth. Recently, misconceptions about their use and misinterpretation of their meaning have even triggered a statement from the American Statistical Association …

Fundamental Analysis of Technology Stocks

This scatterplot uses t-SNE analysis to visualize the relationships between financial performance and market pricing of public technology companies with market capitalizations above $10bil. Select from the various industry groupings at the bottom to explore relationships between different types of businesses. While t-SNE is not technically …

Chris Conlan Appears on Stocks & Jocks

Listen to Chris Conlan on Chicago's biggest finance radio show, Stocks and Jocks, discussing his new book, Algorithmic Trading with Python. Listen to the show here: Stocks and Jocks: Algo Trading.