Measuring Time Series Predictability Using Support Vector Regression (original) (raw)
Related papers
Comparison of Time Series ARIMA Model and Support Vector Regression
International Journal of Hybrid Information Technology, 2020
As one of the most important and costly functions of any business, sales analytics has been the target of many studies for some time now. Knowing and tracking the sales of a business proves useful in all data-driven decisions made from inventory management to shelf layouts in a supermarket. However, forecasting sales rely heavily on data and algorithms strong enough to handle unseen data. Since sales data are in nature time series datasets one of such predictive methods is time series analytics. In this paper, the ARIMA modeling to the seasonality of the data is compared with a machine learning technique, support vector regression. These comparisons are carried out on three different and unrelated datasets and these algorithms' errors when predicting future sales are compared. The results obtained from our analysis show poor results in general due to datasets having large numbers of oscillation and outliers, but for comparison purposes these datasets and results are fine. We conclude that support vector regression produces better results in comparison with time series analytics on all datasets used in this paper.
2006
Recently, novel learning algorithms such as Support Vector Regression (SVR) and Neural Networks (NN) have received increasing attention in forecasting and time series prediction, offering attractive theoretical properties and successful applications in several real world problem domains. Commonly, time series are composed of the combination of regular and irregular patterns such as trends and cycles, seasonal variations, level shifts, outliers or pulses and structural breaks, among others.
Theoretical and practical approaches for time series prediction
2015
The goal of this paper is to discuss two different modern approaches for modeling and prediction of time series – general regression neural networks and support vector regression. It is known that the performances of different approaches from machine learning field are strongly dependent on data. We apply and evaluate our methods on eight different real meteorological series. In order to increase the SVR performances we develop a method for obtaining a SVR optimal multiple kernel.
Mathematics and Statistics, 2023
Univariate time series forecasting is a crucial machine learning issue across many fields notably sentiment analysis, economy, medicine, agriculture, and finance. In this working paper, we tackled comparing the Support Vector Regression (SVR) to the traditional Autoregressive Integrated Moving Average (ARIMA) algorithms in terms of forecasting through a real case study. In fact, the data set used in this investigation has been extracted from the World Bank. The target time series is the American Foreign direct investment, net outflows (% of GDP) which includes the data for 50 years from 1972 to 2021. For analytical and comparison purposes, all the compilations have been done using the R programming language for Windows 10. The statistical findings revealed that, in short-term prediction, the forecast accuracy of both algorithms reduces in terms of error accuracy, significantly. Comparatively, the analysis conducted in this investigation demonstrates that the machine learning algorithms, especially the SVM one perform better than the ARIMA in short-term forecasting since its accuracy functions are the lowest. Thus, we highly recommend future research to compare the advanced machine learning algorithms especially the recurrent neural network algorithms with the classical algorithms, especially with the ARIMA approach in order to choose the best algorithm in terms of results and predictive performance.
Parameter Sensitivity of Support Vector Regression and Neural Networks for Forecasting
Int. Conf. on Data Mining, 2006
Support Vector Regression (SVR) and artificial Neural Networks (NN) promise attractive features for time series forecasting. Despite their attractive theoretical properties, limited empirical studies using small or unbalanced parameter setups yield inconsistent results regarding their empirical accuracy. This paper investigates the accuracy of different configurations of NN and SVR parameters, paying particular attention to the common SVR kernels of polynomial,
2006
Networks (NN) have been successfully applied to forecasting and time series prediction. While conventional statistical methods require specific data preprocessing prior to the forecasting step both, SVR as well as NN need less efforts for the respective tasks due to their theoretical properties. On the other hand, it is known that preprocessing affects performance of classifiers built using these methods.
Prediction and Simulation Spatio-Temporal Support Vector Regression for Nonlinear Data
2020
Spatio-temporal model forecasting method is a forecasting model that combines forecasting with a function of time and space. This method is expected to be able to answer the challenge to produce more accurate and representative forecasting. Using the ability of method Support Vector Regression in dealing with data that is mostly patterned non-linear premises n adding a spatial element in the model of forecasting in the form of a model forecasting Spatio- Temporal. Some simulations have done with generating data that follows the Threshold Autoregressive model. The models are correlated into spatial points generated by several sampling methods. Simulation models are generated to comparing the accuracy between model Spatio-Temporal Support Vector Regression and model ARIMA based on Mean Error, Mean Average Error, Root Mean Square Error, and Mean Average Percentage Error. Based on the evaluation results, it is shown that forecasting with the Spatio-Temporal Support Vector Regressi...
Time series prediction using ls-svms
2008
This paper describes the use of LS-SVMs as an estimation technique in the context of the time series prediction competition of ESTSP 2008 (Finland). Given three different time series, a model is estimated for each series, and subsequent simulations of several points after the last available sample are produced. For the first series, a NARX model is formulated after a careful selection of the relevant lags of inputs and outputs. The second and third series show cyclical or seasonal patterns. Series 2 is modelled by adding deterministic "calendar" variables into the nonlinear regression. Series 3 is first cleaned from the seasonal patterns, and a NAR model is estimated using LS-SVM on the deseasonalized series. In all cases, hyperparameters selection and input selection are made on a cross-validation basis.
Hydrological Processes, 2010
In the recent past, a variety of statistical and other modelling approaches have been developed to capture the properties of hydrological time series for their reliable prediction. However, the extent of complexity hinders the applicability of such traditional models in many cases. Kernel-based machine learning approaches have been found to be more popular due to their inherent advantages over traditional modelling techniques including artificial neural networks(ANNs ). In this paper, a kernel-based learning approach is investigated for its suitability to capture the monthly variation of streamflow time series. Its performance is compared with that of the traditional approaches. Support vector machines (SVMs) are one such kernelbased algorithm that has given promising results in hydrology and associated areas. In this paper, the application of SVMs to regression problems, known as support vector regression (SVR), is presented to predict the monthly streamflow of the Mahanadi River in the state of Orissa, India. The results obtained are compared against the results derived from the traditional Box-Jenkins approach. While the correlation coefficient between the observed and predicted streamflows was found to be 0Ð77 in case of SVR, the same for different auto-regressive integrated moving average (ARIMA) models ranges between 0Ð67 and 0Ð69. The superiority of SVR as compared to traditional Box-Jenkins approach is also explained through the feature space representation.
SVM-based Time Series Prediction with Nonlinear Dynamics Methods
A key problem in time series prediction using autoregressive models is to fix the model order, namely the number of past samples required to model the time series adequately. The estimation of the model order using cross-validation is a long process. In this paper we explore faster alternative to cross-validation, based on nonlinear dynamics methods, namely Grassberger-Procaccia, Kégl and False Nearest Neighbors algorithms. Once the model order is obtained, it is used to carry out the prediction, performed by a SVM. Experiments on three real data time series show that nonlinear dynamics methods have performances very close to the cross-validation ones.