Markus Pelger, Assistant Professor, Stanford AFTLab
Factor Investing using Penalized Principal Components
The fundamental insight from asset pricing theory is that the differences in prices should be explained by exposure to systematic risk factors. Finding the “right” factors has become the central question of asset pricing, that has huge practical importance for investment decisions. Risk factors represent highly profitable investment strategies and can identify under-priced assets. As several hundred variables have been identified as candidate factors, the crucial question is, which risk factors are really important and which factors are subsumed by others? My new statistical methods bring order into the chaos of factors and finds the factors that are relevant for investment and pricing. My estimation approach is a generalization of the widely utilized Principal Component Analysis (PCA). My estimator, Risk Premium PCA (RP-PCA), can be interpreted as PCA generalized with a penalty term to account for the pricing error. Searching for factors explaining asset prices not only co-movement in the data is very important, as factors that can only explain the co-movements do not seem to capture well the risk-return tradeoff. I derive the statistical properties of the new estimator and show that my estimator can find asset-pricing factors, which cannot be detected with PCA, even if a large amount of data is available. My new factors have Sharpe-ratios that are out-of-sample multiple times larger than those of conventional PCA or popular factor models.
Biography
Markus Pelger is an Assistant Professor at the Management Science & Engineering Department at Stanford University and a Reid and Polly Anderson Faculty Fellow at Stanford University.
His research interests are in statistics, financial econometrics, asset pricing and risk management. His work includes contributions in statistical factor analysis, high-frequency statistics, credit risk modeling and management compensation. He is particularly interested in how systematic risk factors and tail risk in the form of jumps influence the price of assets and the incentives of agents. For this purpose he has developed various statistical tools to estimate unknown risk factors from large dimensional data sets and from high-frequency data. He uses them empirically to predict asset prices and construct trading strategies.
Markus received his Ph.D. in Economics from the University of California, Berkeley. He is a scholar of the German National Merit Foundation and he was awarded a Fulbright Scholarship, the Institute for New Economic Thinking and Eliot J. Swan Prize. He has two Diplomas in Mathematics and in Economics, both with highest distinction, from the University of Bonn in Germany.
