Search

The paper proposes a novel hybrid feature selection (FS) method for day-ahead electricity price forecasting. The work presents a novel hybrid FS algorithm for obtaining optimal feature set to gain optimal forecast accuracy. The performance of the proposed forecaster is compared with forecasters based on classification tree and regression tree. A hybrid FS method based on the elitist genetic algorithm (GA) and a tree-based method is applied for FS. Making use of selected features, aperformance test of the forecaster was carried out to establish the usefulness of the proposed approach. By way of analyzing and forecasts for day-ahead electricity prices in the Australian electricity markets, the proposed approach is evaluated and it has been established that, with the selected feature, the proposed forecaster consistently outperforms the forecaster with a larger feature set. The proposed method is simulated in MATLAB and WEKA software.

Times series complexity is a testing standard of a particular algorithm to achieve efficient time execution when it is implemented into programming language. Complexity of the algorithm is differentiated into two parts; those are time series complexity and space complexity. Time series complexity is determined from the numbers of computation steps needed to run algorithm as a function of several data n (input size). However, space series complexity is measured from the memory used by data structure found in the algorithm as the function of several data n. The objective of the study is to test the comparison of three forecasting algorithms among time series complexities by using operational empirical analysis with running time application when the algorithm is used in programming language or a particular application. The algorithms tested in this study were Linear Regression, SMO Regression, and Multilayer Perceptron by using Weka Application. The analysis result and the test result all three algorithms showed that running time was significantly influenced by algorithm's complexity and additional data set numbers that became the input. Furthermore, running time analysis from those three algorithms showed stable rate respectively from SMO Regression, Linear Regression, and Multilayer Perceptron. However, for Big-O algorithm analysis Linear Regression and SMO Regression had almost similar time complexity, but they had far different from Multilayer Perceptron that tended to be more complex. [ABSTRACT FROM AUTHOR]

The paper proposes a novel hybrid feature selection (FS) method for day-ahead electricity price forecasting. The work presents a novel hybrid FS algorithm for obtaining optimal feature set to gain optimal forecast accuracy. The performance of the proposed forecaster is compared with forecasters based on classification tree and regression tree. A hybrid FS method based on the elitist genetic algorithm (GA) and a tree-based method is applied for FS. Making use of selected features, aperformance test of the forecaster was carried out to establish the usefulness of the proposed approach. By way of analyzing and forecasts for day-ahead electricity prices in the Australian electricity markets, the proposed approach is evaluated and it has been established that, with the selected feature, the proposed forecaster consistently outperforms the forecaster with a larger feature set. The proposed method is simulated in MATLAB and WEKA software. [ABSTRACT FROM AUTHOR]