Sina Seyfi

I am a 3rd year PhD student in Finance at Aalto University.

I do research in Empirical Asset Pricing and Big Data. I employ machine learning tools to study the cross-section of stock returns.

Email: sina.seyfi[at]aalto.fi


Research


Essence of the Cross Section

(SSRN)

  • Presentations: Finance Brown Bag seminars (Aalto University), March 2023 / Finance Brown Bag seminars (Copenhagen Business School), June 2023

According to which characteristics do low-mean stocks differ from high-mean stocks? This figure shows how much the distance between characteristics of losers from winner stocks, come from each characteristic. Accounting for 5.7% of the distance, momentum, is the most important characteristic. Moreover, almost 15 characteristics constitute 50% of the distance.

According to which characteristics do low-mean stocks differ from high-mean stocks? This figure shows how much of the distance between characteristics of losers from winner stocks comes from each characteristic. Accounting for 5.7% of the distance, momentum is the most important characteristic. Moreover, almost 15 characteristics constitute 50% of the distance.

Abstract: I develop a method to identify the strongest determinants of expected returns among potentially infinite return predictors. Instead of sorting stocks on characteristics, I sort stocks into portfolios based on their realized returns—the variable of interest—at each month in the past and find the average of each characteristic among assets in each portfolio. Then I create out-of-sample portfolios such that they are as similar as possible to the returns-sorted portfolios regarding 206 characteristics. Differences in characteristics of low- and high-mean stocks determine where the dispersion in expected returns emanates from. I find price-based characteristics are the strongest predictors.


Neighbouring Assets

  • SSRN’s Top Ten download list for: Asset Pricing Models (Topic), European Finance eJournal, Wealth Management eJournal, Decision-Making & Management Science eJournal, DecisionSciRN: Business Analytics (Topic), DecisionSciRN: Predictive Analytics (Sub-Topic), and FinPlanRN: Other Investments (Topic).
  • Presentations: FMA European conference 2023, Aalborg, Denmark, June 2023 / Nordic Finance Network workshop, March 2023, Levi, Finland / Finance Brown Bag seminars (Aalto university), October 2022 / GSF Winter Workshop, Helsinki, November 2022

Abstract:

Firms with similar characteristics display similar expected returns. Defining neighbouring assets as those with the most similar set of characteristics, I show that past returns of an asset’s neighbours predict its future expected returns. If a majority of an asset’s neighbours have performed poorly (well) in the past, it is likely that this asset also performs poorly (well) in the future. By classifying each asset into a decile portfolio based on the past performance of its neighbours, with 94 characteristics, a long-short portfolio generates an out-of-sample annualized Sharpe ratio of 1.15 with a monthly alpha of 2.72% (t = 8.86).

Without loss of generality, if there exist only three firm characteristics, the k closest assets to the asset j (shown as red dots) are defined as the neighbours of asset j. If expected returns are a function of firm characteristics, regardless of the functional form, neighbouring assets display similar expected returns.



Previous Research



Teaching

  • Investments (2022), BSc-level major Finance course
    • Course feedback score 4.13/5 (number of students responded: 42)
  • Investment management (2021), BSc-level major Finance course
    • Course feedback score 4.00/5 (number of students responded: 103)
  • Machine Learning in Finance (2019), MSc elective course
  • Statistics for Finance (2019)
  • Financial Modeling with Python (Workshop) (2019)