Professional Documents
Culture Documents
Ͳ ݔ ܥǡ
IV. SUPPORT VECTOR MACHINES
Ͳ ݔ כ ܥǡ
The support vector machines are structure of the risk
minimization, [14] SVM will seek to minimize an upper ݅ ൌ ͳǡʹǡ ǥ ǡ ܰǤ
bound of the generalization error instead of the empirical error
as in other neural networks. Additionally, the SVMs models
generate the regress function by applying a set of high ݔ andݔ כare call Lagrangian multipliers.They satisfy the
dimensional linear functions. The SVM regression function is equalities,
formulated as follows.
ݔ ݔ כ כൌ Ͳǡ Fig.5 BBL ARIMA Model
(11)
݂ሺݔǡ ߙǡ ߙ כሻ
ൌ ሺ ݔ െ ݔ כሻܭሺݔǡ ݔ ሻ ܾ
ୀଵ
Here, K(x, x) is call the kernel function. The value of the
kernel is equal to the inner product of two vectors ݔ and ݔ in
the feature space ሺݔ ሻ and ሺݔ ሻ , such that ܭ൫ݔ ǡ ݔ ൯ ൌ
ሺݔ ሻ כ൫ݔ ൯ any function that satisfying Mercer’s condition
Fig.6 KBANK SVM Model
can be used as the Kernel function. The Gaussian kernel
function is specific in this study. The SVMs are employing to
estimate the nonlinear behavior of the forecasting data set
because Gaussian kernels tend to give good performance
under general smoothness assumptions.
ܭ൫ݔ ǡ ݔ ൯ ൌ ሺെȁหݔ െ ݔ หȁଶ Ȁሺʹߪ ଶ ሻሻ (12)
V. EXPERIMENTAL RESULT
Fig. 4, Fig. 6, Fig. 8 show us about price prediction of BBL Kbank and SCB
Bank using SVM method for 5 day later.