Abstract
<p>Sweet Potato is a useful plant as a source carbohydrates, proteins, and is used as an animal feed and ingredient industry. Based on data from the Badan Pusat Statistik (BPS), the production fluctuations of the sweet potato in Central Java from year to year are caused by many factor. The production of sweet potato and the factors that affected it if they are described into a pattern of relationships then they do not have a specific pattern and do not follow a particular distribution, such as harvest area, the allocation of subsidized urea fertilizer, and the allocation of subsidized organic fertilizer. Therefore, the production model of sweet potato could be applied into nonparametric regression model. The approach used for nonparametric regression in this study is smoothing spline regression. The method used in regression smoothing spline is generalized cross validation (GCV). The value of the smoothing parameter (λ) is chosen from the minimum GCV value. The results of the study show that the optimum λ value for the factors of harvest area, urea fertilizer and organic fertilizer are 5.57905e-14, 2.51426e-06, and 3.227217e-13 that they result a minimum GCV i.e 2.29272e-21, 1.38391e-16, and 3.46813e-24.</p><p> </p><p><strong>Keywords</strong>: Sweet potato; nonparametric; smoothing spline; generalized cross validation.</p>
Highlights
Indonesia sebagai negara beriklim tropis, memiliki tanah subur dan hasil alam yang beraneka ragam, khususnya di bidang pertanian
Sweet Potato is a useful plant as a source carbohydrates
the production fluctuations of the sweet potato in Central Java from year to year are caused by many factor
Summary
M merupakan orde spline, dan k adalah banyak knot. Dari bentuk matematis fungsi spline, ditunjukkan bahwa spline merupakan model polinomial yang tersegmen (piecewise polynomial), tetapi spline masih bersifat kontinu pada knot-knotnya. 4. Model Regresi Nonparametrik Smoothing Spline Smoothing merupakan suatu proses yang dapat menghilangkan data kasar dengan mengikuti bentuk pola data. Dengan n adalah banyaknya data amatan, λ sebagai parameter pemulus dan f′′(x) adalah fungsi turunan tingkat dua. Dengan fλ adalah estimator dari smoothing spline dan tr(Sλ) ≤ n. Adapun langkah-langkah yang dilakukan untuk menentukan λ pada regresi smoothing spline diasumsikan sebagai berikut. ]n×n menunjukkan submatriks berukuran n × n yang dibentuk dari bagian kiri atas matriks utama [n. Q0t)−1 merupakan submatriks berukuran n × n yang di bentuk dari bagian kiri matriks utama,.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
