Modeling of regional banking activities using spline multiresponse semiparametric regression

Wahyu Wibowo*, Sri Haryatmi, I. Nyoman Budiantara

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


Multiresponse semiparametric regression is statistical model with more than one response and combine parametric regression and nonparametric regression. In this model we used spline function as approach for nonparametric component. Spline estimator in model considered depends on smoothing parameter, so the optimal smoothing parameter must be chosen and for this purposes we modified generalized cross validation criteria. The model will be applied to reveal relationship between variables in analysis of regional economic financial. More specific, the response variables are regional bank activities performance that consists of deposit per capita and credit per capita as well as predictor variables are income per capita, bank density per 1000 population and economic growth. All variables are observed in each Indonesia provinces. Income per capita and bank density per 1000 population variable are treated as parametric predictor variables and economic growth is nonparametric predictor variable. The result shows that the model can manage linear and non linear relationship. Therefore, the model can be applied well with a good performance.

Original languageEnglish
Pages (from-to)391-398
Number of pages8
JournalInternational Journal of Applied Mathematics and Statistics
Issue number14
Publication statusPublished - 2013


  • Credit per capita
  • Multiresponse semiparametric model
  • Saving per capita
  • Smoothing parameter
  • Spline partial


Dive into the research topics of 'Modeling of regional banking activities using spline multiresponse semiparametric regression'. Together they form a unique fingerprint.

Cite this