The SHoF Fund Data from Morningstar is live!

Jan. 25, 2024

The newly launched SHoF Fund Data is an extensive dataset featuring detailed historical records of over 9,000 investment funds available in the Nordic countries. Henrik Talborn, the head of the data center, provides insights into the center’s objectives and contributions to the financial research community.

The SHoF Fund Data offers a comprehensive historical overview of over 9,000 investment funds accessible in Nordic countries like Sweden, Norway, Denmark, and Finland. This dataset predominantly covers open-end mutual funds, alongside a selection of ETFs, hedge funds, money market funds, and closed-end funds. Based on Morningstar's extensive fund data, the dataset showcases fund performance dating back to 1970. 

 

Q&A with Henrik Talborn 

Could you tell us about the origins of the SHoF Data Center? 

Henrik Talborn (HT): The ShoF Research Data Center was founded as an important part of The Swedish House of Finance in 2011. Since that point we have followed our mission to develop and distribute valuable datasets to the benefit of academic research. 

What range of databases does the SHoF Data Center offer? 

HT: We have focused on developing datasets that is of broad interest and useful in many fields of academic research. For instance, we offer the FinBas dataset, which includes adjusted daily asset prices dating back to the inception of the Stockholm stock exchange in 1912, and also covers Norway, Denmark, and Finland. Other datasets include NASDAQ HFT data from 2010, Serrano's accounting data, PAtLink's corporate patent data, and ESG data from Nordic Compass, among others. 

What research gaps did the data center aim to address? 

HT: Where we really contribute is in terms of accessibility and quality. To the academic community we often distribute very expensive datasets free of charge, or for a tenth or less of the commercial price. That is obviously welcome for students and researchers.  

Furthermore, we never compromise data quality. The datasets that are commercially available and intersect with ours are rarely comparable in terms of quality. Quality is indeed very important if researchers are supposed to draw conclusions based on the data. Industry practitioners appear to have higher acceptance for lower quality data while being the data vendors’ most important segment. That leaves us with a job to do. But to create datasets that meet academic standards requires substantial time and effort.   

Can you give some examples? 

HT: One notable example is the NASDAQ HFT dataset. It’s a massive dataset where each day's data is a binary file detailing every event for each instrument. Parsing this data alone could take years, not to mention the additional time required for a researcher to develop software to structure it for research purposes. We've streamlined this process, offering the parsed and structured data for free. 

Another example is FinBas. Adjusting asset prices for corporate actions may seem straightforward, but it's far from it. This data has undergone extensive corrections and adjustments for about 40 years. A key figure in this endeavor is Erik Eklund, who founded FinBas in 1977 at the Stockholm School of Economics and continues to contribute to its development here at SHoF. This dataset, almost handcrafted in its precision, offers unparalleled quality dating back to 1912.