Sequential support vector machines

Freitas, Nando de; Milo, Marta; Clarkson, Peter; Niranjan, Mahesan; Gee, Alan H.

doi:10.1109/nnsp.1999.788120

Cited by 30 publications

(14 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Minimizing will make a function as flat as possible, thus playing the role of controlling the function capacity. The second term is the empirical error measured by the -insensitive loss function (3). This loss function provides the advantage of using sparse data points to represent the designed function (1).…”

Section: Svm For Regreession Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Support vector machine with adaptive parameters in financial time series forecasting

Cao

Tay

2003

IEEE Trans. Neural Netw.

879

344

View full text Add to dashboard Cite

Section: Svm For Regreession Estimationmentioning

confidence: 99%

“…Therefore, the learning algorithm used should take into account this characteristic. Usually, the information provided by the recent data points is given more weight than that provided by the distant data points [3], [4], as in nonstationary financial time series the recent data points could provide more important information than the distant data points.…”

mentioning

confidence: 99%

Support vector machine with adaptive parameters in financial time series forecasting

Cao

Tay

2003

IEEE Trans. Neural Netw.

879

344

View full text Add to dashboard Cite

“…The best results reported in [11] using SVR and KKF are shown for the purpose of comparison. The results obtained for [3] SSVM using noise estimation scheme explained in previous section is also shown. Different window lengths and kernel functions were tried for SSVM and KARX model and best results are reported.…”

Section: Mackey-glass Time Series Predictionmentioning

confidence: 92%

“…A. K. Suykens [4] also falls into this category of static function mapping. The Sequential support vector machine proposed in [3] uses the Kalman recursions for updating the parameters, which was also used by F. N. Chowdhury [12]. In this paper, we estimate the parameters using similar approach.…”

Section: Related Approachesmentioning

confidence: 99%

Kernel Auto-Regressive Model with eXogenous Inputs for Nonlinear Time Series Prediction

Kini

Sekhar

2007

2007 International Conference on Computing: Theory and Applications (ICCTA'07)

View full text Add to dashboard Cite

In this paper we present a novel approach for nonlinear time series prediction using Kernel methods. The kernel methods such as Support Vector Machine(SVM) and Support Vector Regression(SVR) deal with nonlinear problems assuming independent and identically distributed (i.i.d.) data, without explicit notion of time. However, the problem of prediction necessitates temporal information. In this regard, we propose a novel time series modeling technique, Kernel Auto-Regressive model with eXogenous inputs (KARX) and associated estimation methods. Amongst others the advantage of KARX model compared to the widely used Nonlinear Auto-Regressive eXogenous (NARX) model (which is implemented using Artificial Neural Network (ANN)) is, implicit nonlinear mapping and better regularization capability. In this work, we make use of Kalman recursions instead of quadratic programming which is generally used in kernel methods. Also, we employ online estimation schemes for estimating model noise parameters. The efficacy of the approach is demonstrated on artificial time series as well as real world time series acquired from aircraft engines.

show abstract

“…Therefore, it is advantageous to give more weights on the information provided by the recent data than that of the distant data (Freitas, Milo, & Clarkson, 1999). Enlightened by this fact, an innovative SVM approach called C-ascending SVM was proposed by Tay and Cao (2002), where e-insensitive error is multiplied by the exponential weight function to illustrate the different weights from distant data and recent data in time sequence.…”

Section: The Proposed Methodsmentioning

confidence: 99%

Integration of similarity measurement and dynamic SVM for electrically evoked potentials prediction in visual prostheses research

Peng

et al. 2011

Expert Systems with Applications

View full text Add to dashboard Cite

Sequential support vector machines

Cited by 30 publications

References 11 publications

Support vector machine with adaptive parameters in financial time series forecasting

Support vector machine with adaptive parameters in financial time series forecasting

Kernel Auto-Regressive Model with eXogenous Inputs for Nonlinear Time Series Prediction

Integration of similarity measurement and dynamic SVM for electrically evoked potentials prediction in visual prostheses research

Contact Info

Product

Resources

About