The paper refers to multimanager management of Mutual Funds portfolios, i.e. a kind of financial assets directly investing in Mutual Funds referring to a specific market (equities, bonds, corporates, derivatives) instead than in financial stocks. In particular, Mutual Funds portfolios investing on a single asset category are considered. The goal is to provide a method to rank Mutual Funds investment style with respect to the returns of a target index, the so-called benchmark. To this aim, a rolling constrained multiple linear regression model is considered in order to estimate the composition of each Mutual Fund portfolio as well as of the reference index (benchmark), and starting from such compositions, the Mean Integrated Squared Error (MISE) is computed to measure the proximity of each Mutual Fund portfolio returns to the benchmark portfolio returns. A visual inspection of this proximity is also provided using parallel coordinates plot. The method allows to identify an appropriate management style for each Mutual Fund. Its effectiveness is evaluated through an application on a set of Italian Mutual Funds operating in the equity European market.
Model Based Visualization of Portfolio Style Analysis
CONVERSANO, CLAUDIO;
2004-01-01
Abstract
The paper refers to multimanager management of Mutual Funds portfolios, i.e. a kind of financial assets directly investing in Mutual Funds referring to a specific market (equities, bonds, corporates, derivatives) instead than in financial stocks. In particular, Mutual Funds portfolios investing on a single asset category are considered. The goal is to provide a method to rank Mutual Funds investment style with respect to the returns of a target index, the so-called benchmark. To this aim, a rolling constrained multiple linear regression model is considered in order to estimate the composition of each Mutual Fund portfolio as well as of the reference index (benchmark), and starting from such compositions, the Mean Integrated Squared Error (MISE) is computed to measure the proximity of each Mutual Fund portfolio returns to the benchmark portfolio returns. A visual inspection of this proximity is also provided using parallel coordinates plot. The method allows to identify an appropriate management style for each Mutual Fund. Its effectiveness is evaluated through an application on a set of Italian Mutual Funds operating in the equity European market.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
