COPULA REGRESSION AND GENERALIZED LINEAR MODELS: A COMPARATIVE STUDY WITH AN APPLICATION
Walid Elbadawy et al.
What the paper says
Classical linear regression, despite its widespread use and simplicity, often proves inadequate for various applications. The presence of nonlinear relationships and non-normal error distributions necessitates the adoption of alternative modeling approaches. Generalized linear models (GLMs) are a prevalent alternative method, yet they require the response variable to belong to the exponential dispersion models (EDMs) family of distributions. In certain practical scenarios, the response variable may follow a distribution outside of this family, highlighting the significance of copula regression, which does not impose stringent conditions on the probability distribution. Consequently, this paper conducts a comparative study between GLMs and copula regression using real world data. To strictly evaluate the predictive performance and generalizability of the models, a 5-fold cross-validation approach was used. The findings demonstrate that the t-copula regression model yielded superior performance compared to the gamma regression with a log-link function.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.