Introducing Non-Linearities and Interaction Terms in a Conditional Asset Pricing Model
1 online resource (40 pages) : PDF
University of North Carolina at Charlotte
Throughout history of the conditional asset pricing literature the goal has been to find the best possible model to explain what determines a firm’s expected stock return. In Dickson (2015) the variables that prove to be best at explaining a firm’s stock return is book-to-market, market capitalization, gross profitability, investment, short-term reversal, and momentum. The aim of this study is to further examine improvements in Dickson (2015) by changing the functional form and adding interaction terms between the variables. The chosen methodology is a version of the popular Fama-Macbeth regressions which are well documented in the literature to determine the added risk premium associated with firm characteristics. By allowing for the possibility of non-linear characteristics and interaction terms, this study shows that market capitalization follows a significant non-linear relationship with the average stock return and by adding the squared regressor to the model, the explanatory power and risk premium for market capitalization improves. The study further shows that including the interaction between investment and market capitalization improves the explanatory power of the investment variable and the market capitalization variable.
FAMA-MACBETH REGRESSIONFUNCTIONAL FORMINTERACTION TERMSSTOCK-PICKING
Dickson, RobertKirby, ChristopherClark, Steven
Thesis (M.S.)--University of North Carolina at Charlotte, 2016.
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.
Copyright is held by the author unless otherwise indicated.