Upgrade to remove ads
Only $2.99/month

Multivariate Empirical Methods and Performance Persistence

Terms in this set (13)

Eigenvalue
represents the percentage of return variance that is explained by a specific factor
Factor loadings
a vector of values of each asset that determine the sensitivity of each asset for each principal component
Factors analysis
a statistical method that identifies factors and coefficients that optimize a model using explicit assumptions.
Multiple regression models
assumes there is more than one independent variable, and more than one risk factor is used to explain return variances.
Multicollinearity
occurs when two or more independent variables in a regression model are highly correlated with each other.
Two adverse effects of multicollinearity
1. coefficient estimates for the correlated independent variables are inaccurate.

2 the standard errors for the correlated independent variables are overstated.
Stepwise Regression
an iterative process for determining variables that should be included or removed from the regression model.
Nonlinear exposure
occurs when the value of the investment changes based on the size of the change in the factor.
Three dynamic risk exposure models
estimate returns for nonlinear exposures such as market-timing strategies
Conditional correlation
a correlation between two
variables under specified circumstances.
Rolling-window Analysis
another approach for analyzing changing correlations associated with alternative investments.
Style Analysis
focuses on identifying underlying asset returns that can help explain returns for the hedge fund strategy.
Specialized Market Factors
specifically designed to represent the returns to specific hedge fund.
THIS SET IS OFTEN IN FOLDERS WITH...