A modified Unbiased GM(1,1) model

Ji, Peirong; Zhang, Jian; Zou, Hongbo; Zheng, Wei

doi:10.1109/gsis.2011.6044076

Cited by 8 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2.1 Overview of grey models accumulation: GM(1,1) model Among the miscellaneous grey mechanisms, GM(1,1) is significantly used with a certain degree of accuracy. The modelling approach goes through the steps wisely as follows ( Ji and Zhang, 2011;Rathnayaka et al, 2015):…”

Section: Methodsmentioning

confidence: 99%

“…Among the miscellaneous grey mechanisms, GM(1,1) is significantly used with a certain degree of accuracy. The modelling approach goes through the steps wisely as follows (Ji and Zhang, 2011; Rathnayaka et al , 2015): Step 1: develop a conceptual modeling language for through and concepts. Step 2: determine the factors. Step 3: examine the causal relations among the factors which have been identified in Step 2. Step 4: identify the most suitable inputs and outputs to build a dynamic model. Step 5: evaluate the model accuracies and make conclusions.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

An unbiased GM(1,1)-based new hybrid approach for time series forecasting

Rathnayaka

Seneviratna

Jiang

et al. 2016

View full text Add to dashboard Cite

Purpose The time series forecasting is an essential methodology which can be used for analysing time series data in order to extract meaningful statistics based on the information obtained from past and present. These modelling approaches are particularly complicated when the available resources are limited as well as anomalous. The purpose of this paper is to propose a new hybrid forecasting approach based on unbiased GM(1,1) and artificial neural network (UBGM_BPNN) to forecast time series patterns to predict future behaviours. The empirical investigation was conducted by using daily share prices in Colombo Stock Exchange, Sri Lanka. Design/methodology/approach The methodology of this study is running under three main phases as follows. In the first phase, traditional grey operational mechanisms, namely, GM(1,1), unbiased GM(1,1) and nonlinear grey Bernoulli model, are used. In the second phase, the new proposed hybrid approach, namely, UBGM_BPNN was implemented successfully for forecasting short-term predictions under high volatility. In the last stage, to pick out the most suitable model for forecasting with a limited number of observations, three model-accuracy standards were employed. They are mean absolute deviation, mean absolute percentage error and root-mean-square error. Findings The empirical results disclosed that the UNBG_BPNN model gives the minimum error accuracies in both training and testing stages. Furthermore, results indicated that UNBG_BPNN affords the best simulation result than other selected models. Practical implications The authors strongly believe that this study will provide significant contributions to domestic and international policy makers as well as government to open up a new direction to develop investments in the future. Originality/value The new proposed UBGM_BPNN hybrid forecasting methodology is better to handle incomplete, noisy, and uncertain data in both model building and ex post testing stages.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

An unbiased GM(1,1)-based new hybrid approach for time series forecasting

Rathnayaka

Seneviratna

Jiang

et al. 2016

View full text Add to dashboard Cite

show abstract

“…The GM(1, 1) model is one of the best known and most used models belonging to grey systems theory ( Ji et al, 2011) For a better understanding of how this model provides predictions, we start by introducing some definitions and sentences (Liu and Lin, 2010;Zhao et al, 2012):…”

Section: Grey Forecasting -The Gm(11) Modelmentioning

confidence: 99%

GM(1, 1) in bankruptcy forecasting

Delcea

Bradea

Maracine

et al. 2013

Grey Systems: Theory and Application

View full text Add to dashboard Cite

Purpose -The present paper tries to give a new vision on the firm's future evolution forecasting. By taking into account some of the current values of its symptoms and applying one of the most used models in the grey systems theory, namely the GM(1,1), the predictions related to its future symptoms' values can be determined. Having these projected values and the grey economic-financial matrix, K, the future diseases that can hit a company can be depicted along with their causes. The paper aims to discuss these issues. Design/methodology/approach -Forecasting the future development of a firm is always an important issue in firm's survival in nowadays economy. Most of all, it is extremely important to be aware all the time about the inner and outer factors than can make a difference between a successful and a bankrupt firm. For this, here the authors have used three GM(1,1) models for forecasting the future symptoms (expressed through financial indicators) and performance indicator of a firm. Each time, based on the determined accuracy rate, a specific GM model has been chosen for every indicator's forecasting. Findings -Considering some previous researches and findings in bankruptcy modelling and diagnosis, this paper enlarges their applicability by adding the possibility to make future predictions on the indicators' evolution and to observe and to better manage their causes. As it was expected, the GM(1,1) models used for the forecasting of the various time series variables taken into account were different from one case to another, choosing the best-specific model for each variable case conducted to more accurate data-fit, with direct results in the causes hierarchy. Practical implications -By knowing the main causes that determine a certain state in firms' development and understanding them, the manager can action upon them in a manner that can make the difference between a bankrupt and a real successful firm. Originality/value -The paper succeeds in enlarging the view regarding bankruptcy forecasting by adding a dynamic view over the considered variables. If, in most of the cases when facing with financial forecasting, a single model is used for predictions, here the best GM model has been chosen for each variable based on the obtained accuracy rate. The results are concluding.

show abstract

“…However, the GM(1, 1) model has demonstrated a large capacity for error in practical application processes, so many scholars have proposed a series of improved methods. For instance, the accurate calculating formula of the GM(1, 1) model’s background value was derived through the whitening equation (Tan, 2000; Luo et al , 2003; Wang, Dang and Wang, 2008; Wang et al , 2008b, 2009; Yao et al , 2010; Ji et al , 2011). The new calculation formula for the GM(1, 1) model’s background value was deduced by using the non-homogeneous exponential function to fit the accumulated generation sequence, and it breaks through the restrictions of the development coefficient (Wang, Dang and Wang, 2008; Wang et al , 2008a, b; Xue and Wei, 2009).…”

Section: Introductionmentioning

confidence: 99%

Improved GM(1, 1) model based on concave sequences and its applications

Kong

Zhang

2017

View full text Add to dashboard Cite

Purpose The purpose of this paper is to improve the GM(1, 1) model based on concave sequences. Design/methodology/approach First, the restored sequence of the GM(1, 1) model is proved to be convex, and the residual characters of the GM(1, 1) model for concave sequences are analyzed. Second, two symmetry transformations are introduced to transform an original concave sequence into a convex sequence, and then the GM(1, 1) model is established based on the convex sequence. Findings Compared with the traditional modeling method, the new method has high accuracy and is applicable for all concave sequence modeling. Practical implications Two cases are used to illustrate the superiority of this modeling method. Case A is to predict China’s per capita natural gas consumption, and case B is to predict the annual output of an oilfield. Originality/value The application scope of GM (1, 1) model is greatly extended.

show abstract

A modified Unbiased GM(1,1) model

Cited by 8 publications

References 2 publications

An unbiased GM(1,1)-based new hybrid approach for time series forecasting

An unbiased GM(1,1)-based new hybrid approach for time series forecasting

GM(1, 1) in bankruptcy forecasting

Improved GM(1, 1) model based on concave sequences and its applications

Contact Info

Product

Resources

About