## ESG News Sentiment Scores Dataset - Trial

Amenity Analytics has created an ESG dataset that provides company and sector-level analysis of hundreds of news sources for monitoring and tracking ESG issues in-depth.

This trial dataset is an industrial-scale contextual NLP applied to thousands of news sources to develop in-depth, real-time scoring at the company level on ESG issues relating to companies globally. This dataset excludes companies with fewer than one result per day, leaving 1,227 of the largest public US companies to examine. Our sources comprehensively include top international, national, regional, local, and business news along with additional Amenity identified high quality sources. The dataset tracks daily ESG extractions from January 2018 to June 2021.

The Amenity ESG language model employs contextual analysis and language patterns to provide investors with real-time, broadly aggregated, and unbiased environmental-, social- and governance-related evidence and scores. Amenity’s taxonomy consists of 22 events across three dimensions (E, S, and G). These events can be viewed in sum to represent the ESG profile of a given company or analyzed individually. Our approach considers the sentiment, event, and materiality to the company described in the text.

Amenity applies natural language processing and sentiment analysis on news to derive a numerical score (signal). The score is the result of the net sentiment divided by the total negative and positive extractions in the transcript, and per ESG topic. Also includes counts that make up those scores.

The Values in this dataset are represented with scores between -1 and 1. A score of -1 is the most negative and +1 is the most positive. Neutral is 0.

The dataset includes total score + counts and score + counts per key driver (Governance, General, Social and Environmental).

---

## Key Benefits

* Systematically evaluate and quantify the materiality of Environmental, Social and Governance (ESG) factors in the news on companies in your investment universe.

* Unbiased and transparent data. Our ESG model employs contextual analysis and language patterns to capture and analyze all critical events across a rich dataset of thousands of news sources providing investors broadly aggregated and unbiased E, S, and G-related evidence and scores.

* Accuracy at scale. Analyze and monitor developments related to your ESG themes across your investment universe of watchlists, portfolios, and individual equities.

* Via Marketplace: Signal file or dashboard interface. Dashboard features an intuitive user interface that applies an ESG lens to companies within your investment universe. Create sentiment summaries, or segment baskets of equities based on distinct ESG profiles.

---

## Use Cases

* Screening and Idea Generation

* Stock Selection and Relative Stock Selection

* News Surveillance and Monitoring

* Factor Attribution

* Baskets and Trading Structure Creation

* Alpha Generation via Analysis

---

## Verticals

* Asset Managers

* Banks

* Capital Markets

* Insurance

---

## Dataset General Information

Description | Value

----|-----

Update Frequency | Daily

Data Source(s) | LexisNexis

Original Publisher of data | Various News Sources

Geographic coverage | World

Time period coverage | 2018-01-01 to present

Is historical data “point-in-time” | Yes

Data Set(s) Format(s) | CSV

Raw or scraped data | Raw?

Number of companies/brands covered | All US public companies and most public companies throughout the world.

Standard entity identifiers | Ticker + Region, FactSet company id, (ISIN, CIK, SEDOL TBD)

---

## Data Description and Data Dictionary

The file contains the the following fields:

Column Name| Description | Data Type | Example

----|-----

date | news publication date for which the sentiment score and counts were calculated | date |"2020-03-01"

symbologyId | FactSet's unique identifier for a company. Helps in tracking a company in instances where the company’s ticker has changed | string | "000D63-E"

companyName | The company name | string | "Apple"

ticker | The company's ticker symbol | string | XXX

totalCountNegative | Total count of negative extractions for a date | integer | "-5"

totalCountPositive | Total count of positive extractions for a date | integer | "3"

totalCountDailyScore | An unweighted sentiment score for a company. Sentiment score defined as (total Positive extraction - total negative extractions) / (total positive extractions - total negative extractions + 1) | float | "-0.676754"

totalWeightedCountNegative | Total weighted count of negative extractions for a date | integer | "-10"

totalWeightedCountPositive | Total weighted count of positive extractions for a date | integer | "12"

totalWeightedCountDailyScore | A weighted sentiment score for a company. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "-0.676754"

kdEnvironmentalCountNegative | Total count of negative extractions related to Environmental key driver, for a date | integer | "-5"

kdEnvironmentalCountPositive | Total count of positive extractions related to Environmental key driver, for a date | integer | "2"

kdEnvironmentalCountDailyScore | An unweighted sentiment score for a company for extraction related to Environmental key driver only. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.975"

kdEnvironmentalWeightedCountNegative | Total weighted count of negative extractions related to Environmental key driver for a date | integer | "-12"

kdEnvironmentalWeightedCountPositive | Total weighted count of positive extractions related to Environmental key driver for a date | integer | "12"

kdEnvironmentalWeightedCountDailyScore | A weighted sentiment score for a company for extraction related to Environmental key driver only. Sentiment score defined as (total weighted Positive extraction - toatl weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.975"

kdGeneralCountNegative | Total count of negative extractions related to General key driver, for a date | integer | "-5"

kdGeneralCountPositive | Total count of positive extractions related to General key driver, for a date | integer | "-5"

kdGeneralCountDailyScore | A weighted sentiment score for a company for extraction related to XXXX. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.975"

kdGeneralWeightedCountNegative | Total weighted count of negative extractions related to XXX for a date | integer | "-12"

kdGeneralWeightedCountPositive | Total weighted count of positive extractions related to XXX for a date | integer | "12"

kdGeneralWeightedCountDailyScore | A weighted sentiment score for a company for extraction related to XXX. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.975"

kdGovernanceCountNegative | Total count of negative extractions related to Governance key driver, for a date | integer | "-6"

kdGovernanceCountPositive | Total count of positive extractions related to Governance key driver, for a date | integer | "8"

kdGovernanceCountDailyScore | An unweighted sentiment score for a company for extraction related to Governance key driver only. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.16543"

kdGovernanceWeightedCountNegative | Total weighted count of negative extractions related to Governance key driver for a date | integer | "-10"

kdGovernanceWeightedCountPositive | Total weighted count of positive extractions related to Governance key driver for a date | integer | "12"

kdGovernanceWeightedCountDailyScore | A weighted sentiment score for a company for extraction related to Governance key driver only. Sentiment score defined as (total weighted Positive extraction - toatl weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.975"

kdSocialCountNegative | Total count of negative extractions related to Social key driver, for a date | integer | "-6"

kdSocialCountPositive | Total count of positive extractions related to Social key driver, for a date | integer | "4"

kdSocialCountDailyScore | An unweighted sentiment score for a company for extraction related to Social key driver only. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.1654

kdSocialWeightedCountNegative | Total weighted count of negative extractions related to Social key driver for a date | integer | "-10"

kdSocialWeightedCountPositive | Total weighted count of positive extractions related to Social key driver for a date | integer | "12"

kdSocialWeightedCountDailyScore | A weighted sentiment score for a company for extraction related to Governance key driver only. Sentiment score defined as (total weighted Positive extraction - total weighted negative extractions) / (total weighted positive extractions - total weighted negative extractions + 1) | float | "0.975"

---

## Update Frequency

* For trials the data does not update

* Paid subscriber data updates daily

---

##Applications

* What you can do: Develop time series, use for backtesting or fundamental analysis or for company rankings

* What you cannot do: Dive into specific event/extractions counts or see what the text extractions were; in addition this cannot be resold in part or in full as a commercial dataset.

---

## Regulatory and Compliance Information

Portions of the Services (including the Content) may be provided through third-party providers, such as FactSet, LexisNexis, and/or EDGAR, which may impose certain restrictions or additional terms and conditions.

---

## Required Information to Start a Subscription

* EIN number

* # of applications

* # of users

* # of regions

* # companies in coverage

* User(s) email addresses

---

## Need Help?

* If you have questions about our products, contact us using the support information: douglas.hopkins@amenityanalytics.com

---

## About Amenity

We develop cloud-based analytics solutions to help businesses draw actionable insights from text on a massive scale. Fortune 100 companies, hedge funds, financial exchanges, and insurance companies rely on our proprietary NLP technology for use on sources ranging from regulatory filings and earnings call transcripts to news coverage, social media activity, and research reports.

* [Amenity Alpha in ESG White Paper] https://info.amenityanalytics.com/l/700953/2022-04-27/2vm333/700953/1651077048dSQpyskp/Amenity_Alpha_in_ESG_White_Paper.pdf