Risk Factors Analysis

When a company submits its 10K and 10Q reports, they are required by the SEC to report all the risks that may impact the operation of the company, its financial results, or the behavior of its stock price.

The risk factors typically constitute several pages of text in the annual / quarterly reports, and most of them are copied from the previous reports. This makes it challenging to understand the risks the company is most concerned about, and to apply this knowledge towards systematic trading.

TipRanks’ Natural Language Processing and Machine Learning engines scan the 10K and 10Q reports to extract the risks and provide the following data points:

1. A breakdown of risks into 6 major categories. Each category is divided into 4 sub-categories, for a total of 24 sub-categories overall.
2. An indication of which risks have been added / removed or changed in relation to the previous report.

The TipRanks Risk Factors Dataset includes a line item for every 10Q / 10K report submitted by a company, for each 10Q and 10K report, we provide:

- The number of overall risks reported
- The number of risks that have been added / removed / changed or remained unchanged since the last report
- The number of risks that have been added / removed / changed or remained unchanged for each of the 24 subcategories

搜集汇总

背景与挑战

背景概述

该数据集通过自然语言处理技术分析上市公司10K/10Q报告中的风险因素，将其归类为6大类别24个子类，并追踪各风险项相对于前次报告的增减变动情况。数据集提供每份报告中风险总数、新增/移除/变更风险数量等量化指标。