Abstract | ||
---|---|---|
A multivariate methodology based on functional gradient descent to estimate and forecast time-varying expected bond returns is presented and discussed. Backtesting this procedure on US monthly data, empirical evidence of its strong forecasting potential in terms of the accuracy of the predictions is collected. The proposed methodology clearly outperforms the classical univariate analysis used in the literature. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1016/j.csda.2006.07.024 | Computational Statistics & Data Analysis |
Keywords | Field | DocType |
expected bond return,dynamic model,strong forecasting potential,bond return,proposed methodology,classical univariate analysis,functional gradient descent,multivariate methodology,us monthly data,functional gradient descent approach,empirical evidence,garch model,gradient descent,term structure | Bond,Econometrics,Gradient descent,Empirical evidence,Multivariate statistics,Forecasting theory,Multivariate analysis,Statistics,Mathematics | Journal |
Volume | Issue | ISSN |
51 | 4 | Computational Statistics and Data Analysis |
Citations | PageRank | References |
4 | 0.86 | 0 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Francesco Audrino | 1 | 14 | 3.36 |
Giovanni Barone-Adesi | 2 | 10 | 1.57 |