Consistently high returns in private equity has lead to a steady increase in the global totalassets under management during the past few decades. Therefore, the relevancy of investing in private equity is obvious. As an investment class, private equity is much younger thanits public counterpart, which is a big part of the reason why the amount of financial researchand modeling on it is quite confined. Nevertheless, the need for forecasting capabilities for anyinvested party in private equity is still great, and the authors of this thesis set out to delivera model which accurately forecasts expected net asset value development of a private equityportfolio and present a confidence interval for it. Furthermore, the thesis serves to present suchresults conditional on macroeconomic scenarios. The scope of the study includes private equityfunds of various investment classes, namely, small-cap and mid-cap buyout, large-cap buyoutand venture capital and growth equity.To achieve an accurate model, the study is based on data from a credible source and threeseparate models are derived and tested against each other. The three models consist of one using a simple historical mean approach, another is based on theory presented by Takahashi andAlexander (2002) (the TA-model), and the third model (the modified TA-model) comes fromresearch by Buchner, Kaserer and Wagner (2009). The TA-model and the modified TA-modelhave at least one parameter which needs to be optimized. This was done using a conditional leastsquare method, utilizing MATLAB’s tool for solving nonlinear optimization problems, fmincon.Subsequent to the derivation of each model, a statistical test (a p-value test) was completed.This resulted in the TA-model being proved to be the best in forecasting net asset value development of private equity funds (which by extension means it is also the best at projectingthe same for an entire private equity portfolio) and was therefore implemented in further areas. By sorting the data on vintage year of the fund, data sets corresponding to pre-definedmacroeconomic periods could be attained. The TA-model was then fitted on these data setswhich produced meaningful results in regards to net asset value development, conditional onthree different macroeconomic scenarios, early-, mid-, and late market cycle. Next, Monte Carlosimulations were performed by stochastically simulating the distributions of funds in the various investment classes, resulting in confidence intervals of potential outcomes. Ultimately, theresults were applied to a mock portfolio designed by the authors to represent reality in fair way.The results of the study allow for two important conclusions to be drawn. Firstly, the authors areconfident that the thesis delivers a model which forecasts net asset value development of privateequity investments within certain confidence intervals in a good way, thereby fulfilling the aimof the study as accurately as possible, given the scope and limitations of the study. Secondly,the investigation provides solid evidence that the net asset value development of a private equityfund is dependent on what market cycle is prevailing at the time of fund commencement, andhow the development varies between such scenarios. Finally, using insights gained during theinvestigation process, the authors identify some potential areas for future studies.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-205034 |
| Date | January 2024 |
| Creators | Gimbringer, Wilmer, Carlsson, David |
| Publisher | Linköpings universitet, Produktionsekonomi |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0023 seconds