International Journal of Academic and Applied Research (IJAAR)   Year: 2020 | Volume: 4 | Issue: 1 | Page No.: 14-18 Confidence Interval of Multiresponse Semiparametric Regression Model Parameters Using Truncated Spline Lilik Hidayati, Nur Chamidah, I Nyoman Budiantara Abstract: Regression model is one of the statistical methods used to determine the functional relationship between predictor and response variables. Based on the pattern of the relationship regression model can be divided into 3 namely parametric regression, nonparametric regression and semiparametric regression. In statistical inference, parameter estimation consists of point and interval estimations. In this paper, we estimate confidence interval of multiresponse semiparametric regression model parameters based on truncated spline involving inverse variance-covariance of the error weight, then it is implemented to simulation data in case of homoscedasticity. Based on the estimated parameters for both large and small samples with a significant level of 0.05 we get accuracy of 100% for using weight and get accuracy less than 100% for using no weight. It can be concluded that in the case of homoscedasticity, the estimated confidence interval of the semiparametric multiresponse model parameters by using weighted truncated spline is better than by using unweighted truncated spline.