^{1}

^{1}

^{*}

A forecast method of the impact factor trend was given based on grey system theory. Using this method, combined with the top 20 management science journals, the grey system GM (1, 1) model was constructed. The model evaluates and predicts the average impact factor trend of the top 20 management science journals.

Journal Citation Reports (JCR) issued by American Information Research Institute (ISI) is an authoritative system for evaluating journals. This paper is based on JCR database and grey system theory, building a trend analysis method for the average impact factor (IF) of the top 20 management science journals. The results can provide reference for related researchers.

Grey system theory [_{1}, t_{2},∙∙∙, t_{n}}, the corresponding original data sequence is

Setting ?t_{k} = t_{k} − t_{k}_{-1}, when ?t_{k} = const, sequence (1) is equal-space sequence. When ?t_{k} ≠ const, sequence (1) is non-equal-space sequence. One-accumulated generate sequence of original data sequence (1) is x^{ (1)} = {x^{ (1)} (t_{1}), x^{ (1)} (t_{2}), ∙∙∙, x^{ (1)} (t_{n})}, wherein

Calculation formulas for reverting one-accumulated generate sequence to original sequence (1) is

When one-accumulated generate sequence was close to nonhomogeneous exponential law change, the response function was the solution of differential Equation (4).

The solution was

in which unknown constants a and b were uncertain parameters. Discrete response function of (4) was

In Equation (5), k = 2, 3, ∙∙∙, n. To determine uncertain parameters a and b, we could use difference Equation (4).

wherein

with z^{ (1)} (t_{k}) = λx^{(1)} (t_{k})+ (1 − λ)^{ (1)} (t_{k − 1}) smoothing x^{(1)} (t_{k}) of difference Equation (6), we could get difference equation

In above formula, z^{(1)} (t_{k}) was called as background value and l ∈ [0, 1] was called as background parameters. At present, there is still no optimum getter for background parameters l, in order to be used simply and easily, we generally take background parameters for 1/2 in reference [^{T} = (B^{T}B)^{−1}B^{T}Y determined, inside Y = [x^{(0)} (t_{2}), x^{(0)} (t_{3}), ∙∙∙, x^{(0)} (t_{n})]^{T}, and

Substituting obtained parameters a and b into Equation (5), we could get GM (1.1) model of sequence x^{(0)}:

The traditional modeling method of GM (1.1) model had the advantages of simple computation, but its fitting and forecast precision sometimes was poor. Integrating solving parameters and determining boundary value together to discuss in [^{(0)}, then we could call model (8) as rough model. Finish machining of rough model (8) namely rewrote third formulas of rough model (8) as follow, where in, a and b were new uncertain parameters.

By using the modeling method of traditional grey system GM (1.1) model, parameters a could be gotten, then using the accumulated generate sequence and corresponding time series of original sequence again. Substituting the accumulated generate sequence x^{ (1)} = {x^{(1)} (t_{1}), x^{(1)} (t_{2}), ∙∙∙, x^{(1)} (t_{n})} and corresponding time series of original sequence t = {t_{1}, t_{2}, ∙∙∙, t_{n}} into above formula, we could determine uncertain parameters a and b with matrix equation [^{T} = (B^{ T}B)^{−1}B^{ T}Y, inside Y = [ x^{(1)} (t_{1}), x^{(1)} (t_{2}), ∙∙∙, x^{(1)} ( t_{n})]^{T} and

Substituting parameters a and b into Equation (9), we got new GM (1.1) model of original sequence x^{(0)}:

One-inverse accumulated generating above formula, we could get the reducing value of original sequence x^{(0) }

With original data, firstly, modeling GM (1.1) model by using grey system theory; then, on the basis of construction grey system GM (1.1) model, we constructed GM (1.1) model based on information mining.

The grey system GM (1, 1) model based on information mining method as follow, wherein k = 1, 2, ∙∙∙, and n.

From 2004 to 2013 the number of the average impact factor (IF) over time trend graph can be seen that the number of the average impact factor (IF) is gradually slowly increase trend.

Based on GM (1, 1) model (12), we forecasted the number of the average impact factor (IF) form 2014 to 2019. Results showed a slowly increase trend. Specific data are shown in

Time (Year) | IF | Time (Year) | IF |
---|---|---|---|

2004 | 5.8205 | 2009 | 9.2033 |

2005 | 6.77 | 2010 | 9.45755 |

2006 | 7.50235 | 2011 | 12.41825 |

2007 | 7.44335 | 2012 | 11.0115 |

2008 | 8.3025 | 2013 | 12.41825 |

Time (Year) | IF | Time (Year) | IF |
---|---|---|---|

2014 | 13.66167 | 2017 | 17.30375 |

2015 | 14.78141 | 2018 | 18.72200 |

2016 | 15.99293 | 2019 | 20.25650 |

This paper gives a modeling way based on information mining and grey system theory. On the one hand, this way greatly improved GM (1, 1) model’s fitting precision and prediction accuracy; on the other hand, it maintains the advantage of the traditional modeling method which is simple. The grey system GM (1, 1) model of impact factor (IF) trend was constructed. The model evaluates and predicts the average impact factor (IF) trend of the top 20 management science journals. Case analysis verified the validity and usefulness of the information mining method. The results can provide reference for related researchers.

Xiao Dong,Shiqiang Zhang, (2015) Impact Factor Dynamic Forecasting Model for Management Science Journals Based on Grey System Theory. Open Journal of Social Sciences,03,22-25. doi: 10.4236/jss.2015.33005