Wolfe Research Senior Quantitative Analyst, Yin Luo, hosted a webcast discussing text mining of unstructured corporate filing data.
Search Coverage List, Models & Reports
Search Results1-10 out of 19
This month, we highlight one recent academic paper by Löffler, et al , using linguistic tone analysis on Moody’s rating reports. Löffler, et al  research coincides with one of our recent papers (see Rohal, et al ), in which we apply NLP (Natural Language Processing) and machine learning techniques on corporate filing data from the EDGAR database. We are in the process to expand our research to other textual datasets.
The retail industry has been facing unprecedented challenges. Whether it’s Amazon’s tight grip on retail sales known as the “Amazon Effect” or the growth in conversion rates through mobile apps, retailers are addressing these challenges through innovation. Investors are facing similar challenges. How can investors assess and value an industry that is rapidly reshaping? Investors also need innovative tools to help them make investment decisions.
Wolfe Research Managing Director, Quantitative Analysis, Strategy and Economics, and Vice Chairman of Wolfe Research Steve Fleishman alongside midstream analysts Alex Kania and Keith Stanley, hosted a webinar on their initiation of four companies today.
In the first three parts of the QES Handbook of Active
Investing series, we discussed Big Data, factor backtesting, multifactor models, style rotation, machine
learning and our global stock selection models. In this research, we address the practical issues in portfolio
implementation – from risk models, transaction cost analysis, portfolio construction, to performance
measurement and attribution.
Traditionally, investors rely primarily on the sellside consensus for earnings and revenue forecasts. In this research, we study an alternative data source
based on the concept of crowdsourcing. Estimize is an online platform that allows individuals with different
backgrounds to contribute their financial forecast. We find Estimize estimates to be not only more accurate
and timelier than the sell-side, but also highly complementary to traditional factors.
With the rapid development and innovation in computing power and machine learning algorithms, processing unstructured textual information to generate useful numerical signals becomes increasingly important. In this research we take advantage of distributed cloud computing and advanced Natural Language Processing (NLP) algorithms to systematically analyze corporate filings from the EDGAR database – the official corporate filing database maintained by the SEC.
After sifting through over 2,000 papers presented at the annual AEA (American Economic Association) and AFA (American Finance Association)
conference in January 2017 in Chicago, we have carefully chosen around 150 most interesting and relevant
papers in five fields: alpha signal, risk and portfolio construction, multi-asset and global macro, trading, and
ESG for this issue. The AEA/AFA conference is probably the largest and most well-known event for
graduating PhD students seeking jobs in either academia or industry. In our opinion, the papers presented
during the conference represent some of the best and most innovative research.
Ravenous Profiling Undoubtedly, the notion of profiling is controversial. However, when the dialogue centers
on “company profiling”, the debate is more engaging and frankly, kind of interesting. In this novel research, we
analyze a multitude of fundamental and statistical techniques engineered to assess the quality, governance,
and accounting integrity of a company. Essentially, the goal is to devise systematic methods to separate high
quality firms from those that are in La La Land.
This paper forms the third part of our QES Handbook of Active Investing series. In Part I (The Big and the Small Sides of Big Data), we discussed data contents and data science. In Part II (Signal Research and Multifactor Models), we reviewed the methodologies of factor backtesting and multifactor modeling techniques. In this research, we address some of the more advanced topics. In particular, we focus on how to incorporate style rotation/factor timing and machine learning in equity models, and finally introduce the first of our global stock-selection models – the LEAP (L-Economic Alpha Processing).
- 1 of 2
- next →