Invesco Global Thematic Innovation Equity Fund
Capturing innovation earlier

Invesco Global Thematic Innovation Equity Fund

Access innovations earlier through the analytical power of Natural Language Processing (NLP), an application of artificial intelligence.

Key advantages

The fund in 60 seconds

The investment process

The goal is to capture innovations by investing in equities that are associated with multiple investment themes around the three key overarching megatrends.

The Invesco Quantitative Strategies (IQS) team has developed analytical tools through their research in machine learning and Natural Language Processing. These tools enable them to analyse news data with the aim of finding companies that have the highest exposure to and the highest relevance for each investment theme. Or in other words: companies who will profit most from the rising importance of each investment theme.

The investment concerns the acquisition of units in an actively managed fund and not in a given underlying asset.²

What is Natural Language Processing?
What is Natural Language Processing?
HOW NLP IS USED

In investment management, NLP techniques can be used to support investment decisions through:
  • Keyword extraction
     
  • Automatic summarisation
    Topic segmentation
  • Speech recognition
     
  • Sentiment analysis
    Named entity recognition
  • How Invesco
    uses NLP

Natural Language Processing

Natural Language Processing (NLP) is an application in the field of artificial intelligence in which computer algorithms can analyse, understand and derive meaning from human language and text in an automated way.

Scroll and discover NLP
HOW INVESCO USES NLPfor the Invesco Global Thematic Innovation Equity Fund RECOGNITIONNAMED ENTITY I don't know! Ops!!! I love it! HAHAHAH! SENTIMENTANALYSIS SPEECHRECOGNITION A U T O M A T I C S U M M A R I S A T I O N SEGMENTATIONTOPIC AUTOMATICSUMMARISATION KEYWORD EXTRACTION 1 2 3 3. ASSEMBLING THE PORTFOLIO At the end, the identifiedcompanies make up a broadlydiversified multi-themeportfolio. 2. COMP ANY DET E C TION T he t e am u se s NLP algorithms t o s c an millions o f n e w s da t a i t ems on an ongoing b asis t o d e t ect which c om p ani e s a r e men tioned t og e ther with the k e y w o r ds in the new s. Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in KEYWORDEXTRACTION1. KEYWORD EXTRACTION The team uses keywordextraction algorithms toextract innovation-relatedkeywords from documentssuch as academic researchor broker research. Topic segmentationDetect whether differenttopics are discussed in asingle text (e.g. in a longerconversation), and splittext into the appropriatesegments. Keyword ExtractionSwiftly extract relevantkeywords from text data. AutomaticsummarisationSummary generation fromvast amounts of text.For example from,research reports. Speech recognitionIdentification of words andphrases in spokenlanguage. For example, in earnings calls. Sentiment analysisInterpretation andclassification of emotionswithin text data.For example, managementsentiment in earningscalls.
HOW INVESCO USES NLP RECOGNITIONNAMED ENTITY I don‘t know! Ops!!! I love it! HAHAHAH! SENTIMENTANALYSIS SPEECHRECOGNITION A U T O M A T I C S U M M A R I S A T I O N SEGMENTATIONTOPIC AUTOMATICSUMMARISATION KEYWORD EXTRACTION 1 2 3 3. ASSEMBLING THE PORTFOLIO At the end, the identifiedcompanies make up a broadlydiversified multi-themeportfolio. 2. COMPANY DETECTION The team uses NLP algorithms to scanmillions of news data items on an ongoingbasis to detect which companies arementioned together with the keywords inthe news. Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in KEYWORDEXTRACTION1. KEYWORD EXTRACTION The team uses keywordextraction algorithms toextract innovation-relatedkeywords from documentssuch as academic researchor broker research. Keyword ExtractionSwiftly extract relevantkeywords from text data. Sentiment analysisInterpretation andclassification of emotionswithin text data.For example, managementsentiment in earningscalls. AutomaticsummarisationSummary generation fromvast amounts of text.For example from,research reports. Topic segmentationDetect whether differenttopics are discussed in asingle text (e.g. in a longerconversation), and splittext into the appropriatesegments. Speech recognitionIdentification of words andphrases in spokenlanguage. For example, in earnings calls.

1. Spotting themes

The investment process starts with the systematic analysis of innovation-related documents such as academic papers, financial research and think tanks' and futurists' publications. With Natural Language Processing algorithms, the team identifies investment themes that underlie the three key megatrends. Every investment theme is a proprietary dictionary of associated, innovation-related keywords that the IQS team have extracted from the documents. Continuously analysing a broad set of documents makes sure that as soon as a new investment theme emerges, it is captured in the fund. The outcome of this process is a set of investment themes.
Spotting themes

2. Identifying companies

In a second step, the team uses NLP algorithms and the theme dictionaries to scan millions of news data items. The goal: to identify the companies with the highest exposure to, as well as the highest relevance for, the investment themes. Why analyse news data? The team believes that the more often a company is mentioned in the news in the context of an investment theme and its keywords, the higher the company’s relevance for the theme is. Hence, it will profit more from the rising importance of the investment theme.
Identifying companies

3. Creating a multi-theme portfolio

At the end, the identified companies make up a broadly diversified multi-theme portfolio where the theme weights and the underlying company weights within the themes are driven by their number of occurrences in the news data.
Creating a multi-theme portfolio

The performance

The IQS team has been testing its investment approach for five years now.

Past performance does not predict future returns.

Cumulative performance

12 months rolling returns (%)

  30/11/2016
30/11/2017
30/11/2017
30/11/2018
30/11/2018
30/11/2019
30/11/2019
30/11/2020
30/11/2020
30/11/2021
Fund - - - - 15.19
Benchmark - - - - 19.27



The theme dictionaries as of 31 December 2021 were used to screen monthly news data for companies that are associated with the investment themes. Performance results do not reflect the deduction of investment advisory fees. A client’s actual return will be reduced by the advisory fees and any other expenses which may be incurred in the management of an investment advisory account.

The IQS team

The Invesco Quantitative Strategies (IQS) team has been working with large amounts of data for over three decades. They have a long track record in in systematically translating large data sets into investable portfolios. Current research projects include: Artificial Intelligence, ESG, Machine Learning and Natural Language Processing.

The team behind this fund

Footnotes


  • Source: Statista as of April 2020.

     

    The investment concerns the acquisition of units in a fund and not in a given underlying asset.

    The Fund is actively managed and is not constrained by its benchmark, the MSCI All Country World Index, which is used for comparison purposes. However, the majority of the Fund’s holdings are likely to be components of the benchmark. As an actively managed fund, this overlap will change and this statement may be updated from time to time.

     

    3 Source: Invesco as of 31 December 2021.

Important information

  • Where individuals or the business have expressed opinions, they are based on current market conditions, they may differ from those of other investment professionals and are subject to change without notice.

     

    For more information on our funds and the relevant risks, please refer to the share class-specific Key Information Documents (available in local language), the Annual or Interim Reports, the Prospectus, and constituent documents, available from www.invesco.eu. A summary of investor rights is available in English from www.invescomanagementcompany.lu. The management company may terminate marketing arrangements. This marketing document is not an invitation to subscribe for shares in the fund and is by way of information only, it should not be considered financial advice. This does not constitute an offer or solicitation by anyone in any jurisdiction in which such an offer is not authorised or to any person to whom it is unlawful to make such an offer or solicitation. Persons interested in acquiring the fund should inform themselves as to (i) the legal requirements in the countries of their nationality, residence, ordinary residence or domicile; (ii) any foreign exchange controls and (iii) any relevant tax consequences. As with all investments, there are associated risks. This document is by way of information only. Asset management services are provided by Invesco in accordance with appropriate local legislation and regulations. The fund is available only in jurisdictions where its promotion and sale is permitted. Not all share classes of this fund may be available for public sale in all jurisdictions and not all share classes are the same nor do they necessarily suit every investor. Fee structure and minimum investment levels may vary dependent on share class chosen. Please check the most recent version of the fund prospectus in relation to the criteria for the individual share classes and contact your local Invesco office for full details of the fund registration status in your jurisdiction. This fund is domiciled in Luxembourg.