Tham khảo tài liệu 'forecasting, structural time series models and the kalman filter by a. c. harvey_4', tài chính - ngân hàng, tài chính doanh nghiệp phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | 170 Stochastic Filtering with Applications in Finance determined by the same factors as credit risk. This is the first study to directly estimate credit spread factors using the state-space representation of a multifactor Vasicek model. By making no prior assumptions about the risk variables driving credit spreads the approach provides a contrast to existing empirical literature and an independent test of theory. Multifactor Model of Credit Spreads 171 Table Parameter Estimates of the One Two and Three Factor Models. April 1996 rMarch 2003 April 1996 rMarch 2008 1 Factor 2 Factors 3 Factors 1 Factor 2 Facto s 3 Fact Log L AIC BIC Kappa 1 Kappa 2 Kappa 3 Theta 1 Theta 2 Theta 3 Sigma 1 Sigma 2 Sigma 3 I The table shows the maximum-likelihood estimates for each of the three parameter q s of each factor under the one two and three-factor models. The Log-L is the maximized value of the log likelihood function. This is also used to calculate the Akaike Information Criterion AIC and Bayesian Information Criterion BIC standard errors based on the inverse Hessian matrix are shown below the parameter estimates. 172 Stochastic Filtering with Applications in Finance Table Credit Spread Summary Statistics and Model Fit Errors. Spread Statistics Error Statistics ADF Stats . p-values Index Avg. bp SD bp Avg. bp MAPE SD bp No Trer id Trend AAA I AA 1 A1