Application of Epanechnikov kernel smoothing technique in disability data

Statistical data contains noise. Smoothing is used to smooth out these noises and present the data as a meaningful one. Kernel methods are nonparametric smoothing tools that can reveal structural features in the data which may not be possible with a parametric approach. This paper applies Epanechnikov kernel method of data smoothing to smooth out the dropout rates of the children with disabilities in the special educational institutions. The continuation probabilities and dropout rates of these children in the special educational institutions are indicators of effectiveness of such education systems. The dropout rates before and after smoothing are graphically presented. The distributions of the crude and smoothed rate are examined. It has been observed that under chi-squared test the smoothed data follows log logistic distribution while the crude data follows triangular distribution.