Erez Katz, Lucena Research CEO and Co-founder
Imagine you have a bird’s-eye view of the entire US consumer credit health. The value that this type of information can bring to an astute macro or US equity investor is substantial. For example, you can tell:
- - How many new loans such as mortgages, home equity lines of credit, auto loans, or student loans have been issued
- - How many new credit cards were opened by sector
- - How many consumers are paying their credit cards in full
- - The average balance on credit cards, average credit utilization, average default rate, etc.
Consumer Credit Analysis
When discussing consumer credit information, it’s important to distinguish between PII (personal identifier information) and trend analysis, which is what this particular data represents. Individual records are obfuscated and aggregated to form trends which in turn are analyzed in order to identify shifts in sentiment based on aggregated consumer behavior.
For years, consumer credit has been one of the main pillars of America’s economic state. The data has been used by many sophisticated investors as a directional predictor for the financial markets’ outlook.
Aggregating and trending the debt amassed by the US consumer in the context of: credit and retail cards, student and auto loans, mortgage and home equity lines of credit and other non-traditional debt can provide an unprecedented foresight into certain consumer sectors. Subsequently, this type of data can be used to drive more informed investment decisions, as we discussed during our Cracking the Code panel event.
Lucena Partnered with Equifax® to Analyze Consumer Credit Trends
The FRED (Federal Reserve Economic Data) department provides a monthly report of consumer credit data usage. The sheer volume of such data in varying formats and update frequencies presents key challenges for reliable consumption.
Lucena Research teamed up with Equifax® with a clear goal in mind: to mobilize Equifax’s six plus terabytes of historical data into actionable investment deliverables using advanced data science and machine learning technology.
Lucena’s goal was to generate several derived offerings for those unable to consume and analyze such a vast amount of data but are interested in integrating consumer credit insights into their investment research.
Lucena focused on three main sectors
- - Homebuilders – direct correlation to mortgages, home equity line of credits, and student loans.
- - Automotive – direct correlation to auto loans and overall consumer credit health.
- - Retail/Consumer Discretionary – direct correlation to retail credit card and traditional credit card spending levels.
After the data validation and feature engineering processes, we collaborated with Equifax to create three products derived from the underlying consumer credit data.
Key Performance Indicators are the fundamental data points from which analysts and subsequently portfolio managers devise their investment thesis. For example, identifying a retailer’s year-over-year growth in sales and comparing such value against street consensus, could present a substantial advantage to the astute investor.
Lucena conducted multi-phase research to validate Equifax’s data and illustrate its predictive power in forecasting KPIs for certain publicly traded constituents. The KPI data feed is a daily report which contains graphical and statistical representations of the effectiveness of Equifax’s consumer credit data in forecasting future quarter-over-quarter growth in revenue.
What has been most exciting is the accuracy of our KPI forecast against the street consensus. In all of the cases above, the Equifax/Lucena KPI forecast beat the average street consensus a rate of 57% to 63%. Most importantly the Equifax/Lucena model was able to identify anomalies much more effectively than the street projected. Subscribing to the KPI feed will give you access to a daily feed sent directly to your email every morning.
Model Portfolios are simulations of algorithmically driven paper traded portfolios. In essence, Lucena’s platform carries forward the very same rules of its backtest in a perpetually traded simulation to ensure no overfitting and, in general, the authenticity of the machine learning model.
Lucena’s machine learning classifier relies on Equifax’s features to gain insight into the homebuilders and retail sectors. The multi-factor model portfolios are available exclusively on Lucena’s platform and consist of buy/sell signals predicated on extensive feature engineering, feature selection, and advanced machine learning technology.
Smart Data Feed
For those advanced researchers who are interested in combining the Equifax signals with their internal data and models, we’ve created the Smart Data Feeds. It is a simple 12 factor file produced daily consisting of an ordered list of constituents based on the signal’s strength.
Each constituent is marked with buy/sell/do nothing signals, each signal’s confidence scores, and other statistical scores all pointing to the signal strength. The models that produce these daily signals are also derived from Equifax’s consumer credit features combined with Lucena’s global macro, technical, and fundamental features.
These signals are also predicated on advanced machine learning techniques which are set to generate multi-factor models customized to Equifax’s specific asset universe, mainly automotive, homebuilders, and retail.
Consumers of the Smart Data Feeds Can:
- - Trade the signal: Buy/sell/do nothing signals can be used as triggers for entry into long/short or exit.
- - Use the machine learning model’s output: Using Lucena’s scores and ranks, users can apply rules for entry based on values above certain thresholds. For example, enter all positions with a rank of 5 and above and a confidence score above 75%. Lucena’s platform, QuantDesk, allows users to backtest and validate various considerations.
Our partnership with Equifax is already bearing fruit. The KPI forecasting for example, year-over-year revenue growth forecasts, have consistently outperformed the street consensus by a margin of 2 to 1, or 63% with some of the constituents we cover.
We continue to witness a strong market appetite for easily accessible and actionable information derived from big data such as Equifax’s consumer credit data. Equifax’s consumer credit data is well suited for KPI forecasting as well as global macro, sector, and individual securities sentiment analysis especially when such data is combined with other complementary data sets. In the coming months we will share additional details of our results and host a webinar to answer your questions live.
In the meantime, if you have any questions or are interested in custom research contact us.
DISCLAIMER PERTAINING TO CONTENT DELIVERED & INVESTMENT ADVICE: This information has been prepared by Lucena Research Inc. and is intended for informational purposes only. This information should not be construed as investment, legal and/or tax advice. Additionally, this content is not intended as an offer to sell or a solicitation of any investment product or service. Please note: Lucena is a technology company and neither manages funds nor functions as an investment advisor. Do not take the opinions expressed explicitly or implicitly in this communication as investment advice. The opinions expressed are of the author and are based on statistical forecasting on historical data analysis. Past performance does not guarantee future success. In addition, the assumptions and the historical data based on which opinions are made could be faulty. All results and analyses expressed are hypothetical and are NOT guaranteed. All Trading involves substantial risk. Leverage Trading has large potential reward but also large potential risk. Never trade with money you cannot afford to lose. If you are neither a registered nor a certified investment professional this information is not intended for you. Please consult a registered or a certified investment advisor before risking any capital. The performance results for active portfolios following the screen presented here will differ from the performance contained in this report for a variety of reasons, including differences related to incurring transaction costs and/or investment advisory fees, as well as differences in the time and price that securities were acquired and disposed of, and differences in the weighting of such securities. The performance results for individuals following the strategy could also differ based on differences in treatment of dividends received, including the amount received and whether and when such dividends were reinvested. Historical performance can be revisited to correct errors or anomalies and ensure it most accurately reflects the performance of the strategy.
Copyright © 2019, Equifax Inc., Atlanta, Georgia. All rights reserved. Equifax is a registered trademark of Equifax Inc