Thursday, October 3, 2013

Fuzzy Interation Regression for Short Term Load Forecasting

It's a common but embarrassing practice that many researchers cite some papers without even reading them. As a result, more and more papers are being generated and cited without improving the state of the art. When I was reading the 1200 papers for my PhD, I first started with the papers with high citation count. However, I was shocked about the low quality of some frequently cited papers. After reading many of them, I realized that the citation count may not really imply the quality. An outcome of the PhD study is the capability of critical thinking. Performing critical review of the literature is one way to achieve this goal.

The paper below proposes a fuzzy interaction regression approach to short term load forecasting. We pointed out some significant issues of a highly cited paper on the similar subject. We submitted this paper to a top journal in the fuzzy optimization field, so that our approach can be reviewed by experts in the fuzzy community. We used ISO New England data in this paper, because the data is publicly available for the researchers who are interested in reproducing our results. This is probably the most math-intensive paper I wrote. Ignoring the equations will not affect the understanding of our approach at conceptual level.

The paper is available for download on Springer Link.

Citation
Tao Hong and Pu Wang, "Fuzzy interaction regression for short term load forecasting", Fuzzy Optimization and Decision Making, vol.13, no.1, pp. 91-103, March, 2014

Fuzzy Interaction Regression for Short Term Load Forecasting

Tao Hong and Pu Wang
Abstract

Electric load forecasting is a fundamental business process and well-established analytical problem in the utility industry. Due to various characteristics of electricity demand series and the business needs, electric load forecasting is a classical textbook example and popular application field in the forecasting community. During the past 30 plus years, many statistical and artificial intelligence techniques have been applied to short term load forecasting (STLF) with varying degrees of success. Although fuzzy regression has been tried for STLF for about a decade, most research work is still focused at the theoretical level, leaving little value for practical applications. A primary reason is that inadequate attention has been paid to the improvement of the underlying linear model. This application-oriented paper proposes a fuzzy interaction regression approach to STLF. Through comparisons to three models (two fuzzy regression models and one multiple linear regression model) without interaction effects, the proposed approach shows superior performance over its counterparts. This paper also offers critical comments to a notable but questionable paper in this field. Finally, tips for practicing forecasting using fuzzy regression are discussed.

Update: 2/16/2014
I'm surprised and pleased to know that my paper is now the most popular paper of Fuzzy Optimization and Decision Making (see the screen shot below)!


No comments:

Post a Comment

Note that you may link to your LinkedIn profile if you choose Name/URL option.