DEA Models available in MaxDEA X

MaxDEA X包含的DEA模型

以下列出 MaxDEA X 支持的全部模型与方法。

如果您在列表中没有找到需要的模型,请联系我们:MaxDEA@maxdea.com

1

Model Type模型分类

  1. Envelopment Model (包络模型)
  2. Multiplier Model and FDEF-based Multiplier Model (乘数模型和基于FDEF的乘数模型,即前沿生产函数模型: vx - uy + u₀ = 0)
  3. FDH 模型
2

Distance to measure efficiency用于测量效率的距离类型

  1. Radial (CCR 1978; BCC 1984) (径向距离)
  2. Maximum Distance to Frontier (ERM, Enhanced Russel Measure, Pastor, Ruiz, and Sirvent 1999; SBM, Slacks-based Measure, Tone 2001) (至强有效前沿的最远距离,即SBM模型)
  3. Minimum Distance to Weak Efficient Frontier (Charnes, Roussea, and Semple1996) (至弱有效前沿的最近距离)
  4. Minimum Distance to Strong Efficient Frontier (Closest Target), with full features: strong monotonicity algorithm; CRS, VRS, NIRS and NDRS; Non-oriented, Input-oriented and Output-oriented. (至强有效前沿的最近距离,MinDS距离)

    Ref. to(J. Aparicio et al., 2007; G. R. Jahanshahloo et al., 2012; Gholam Reza Jahanshahloo, Roshdi, & Davtalab-Olyaie, 2013; Olyaie, Roshdi, Jahanshahloo, & Asgharian, 2014; J Aparicio et al., 2017; Zhu et al., 2018; Zhu et al., 2022)

  5. Directional Distance Function (Chambers, Chung, and Färe 1996; Chung, Färe, and Grosskopf 1997) (方向距离函数)

    Direction Vector can be (方向向量类型)

    • ( -|x0|, |y0|, -|b0| )'
    • ( -|x̅|, |y̅|, -|b̅| )'
    • Vector (1, 1, ..., 1)'
    • Range (RDM, Portela, Thanassoulis, and Simpson 2004)
    • Customized (same for all DMUs) (为所有DMU自定义相同的方向向量)
    • Customized (DMU specific) (为各DMU自定义不同的方向向量)
  6. A Series of Weighted Additive Models (加权加性距离)
    • Simple Additive model: Weights = (1, 1, 1, ...) (不加权)
    • Normalized Weighted Additive (Lovell and Pastor 1995) (标准化权重)
    • Weights = 1/|x0|, 1/|y0|
    • Weights = 1/|x̅|, 1/|y̅|
    • Range Adjusted Measure (RAM, Cooper, Park, and Pastor 1999) (RAM模型)
    • Bounded Adjusted Measure (BAM, Cooper, Pastor, Borras, Aparicio, and Pastor 2011) (BAM模型)
    • Directional Slacks-based Measure (DSBM, Fukuyama and Weber 2009) (定向SBM模型)
    • Customized Weights (same for all DMUs) (为各DMU自定义相同的权重)
    • Customized Weights (DMU specific) (为各DMU自定义不同的权重,使用此类型实现一般化的“非径向方向距离函数”,即NDDF模型)
  7. Hybrid Distance(Radial and SBM Measure): (EBM, Epsilon-based Meaure,Tone and Tsutsui 2010) (EBM模型)
  8. Cost (成本效率)
  9. Revenue (收益效率)
3

Orientation to measure efficiency模型导向

  1. Input-oriented (投入导向)
  2. Output-oriented (投入导向)
  3. Non-oriented (非导向)
4

RTS to measure efficiency规模报酬

  1. Constant returns to scale (CRS) (规模报酬不变)
  2. Variable returns to scale (VRS) (规模报酬可变)
  3. Non-increasing returns to scale (NIRS) (规模报酬非增)
  4. Non-decreasing returns to scale (NDRS) (规模报酬非减)
  5. Decomposition of Efficiency or TFP Index (效率或生产率指数分解)
5

TFP Index: Malmquist Index and Hicks-Moorsteen Index (also called HMB Index)全要素生产率指数: Malmquist指数和HMB指数

  1. Adjacent Malmquist (相邻参比)
  2. Fixed Malmquist (固定参比)
  3. Global Malmquist (全局参比)
  4. Sequential Malmquist (序列参比)
  5. Window-Malmquist (Adjacent) (相邻窗口参比)
  6. Window-Malmquist (Fixed) (固定窗口参比)
  7. Global with Sequential Malmquist (全局和序列参比)
TFP index decomposition: Efficiency Change (catch-up), Technological Change (frontier shift), Scale Efficiency Change, biased Technological Change, TC=OBTC*IBTC*MATC (Fare et al 1997)
Three types of indices can be computed: Index(t-1, t); Index (t-n, t) : n is user-defined; Index (t0, t): t0 is the initial period.
6

Window DEA窗口DEA

7

Cluster model群组参比模型,广义DEA模型

  1. Self-benchmarking (自我参比)
  2. Cross-benchmarking (交叉参比)
  3. Downward-benchmarking (向下参比)
  4. Upward-benchmarking (向上参比)
  5. Lower-adjacent-benchmarking (下方临群参比)
  6. Upper-adjacent-benchmarking (上方临群参比)
  7. Window-benchmarking (窗口参比)
  8. Fixed-benchmarking (固定参比)
8

Other models

  1. Super-efficiency model (超效率模型)
  2. Modified SBM (Sharp et al 2007) (MSBM模型)
  3. Modified SBM (Lin et al 2019) (MSBM模型)
  4. Cross efficiency model (交叉效率模型)

    Second-stage methods are available (第2阶段方法):

    Minimize/Maximize the trade balance of other DMUs as a whole

    • Blanket Benevolent (Type I in Doyle and Green 1995)
    • Blanket Aggressive (Type I in Doyle and Green 1995)

    Maximize/Minimize the cross-efficiency of other DMUs as a whole

    • Blanket Benevolent (Type II in Doyle and Green 1995)
    • Blanket Aggressive (Type II in Doyle and Green 1995)

    Maximize/Minimize the cross-efficiency of other DMUs as a whole

    • Blanket Benevolent (Ruiz (2013))
    • Blanket Aggressive (Ruiz (2013))

    Maximize/Minimize the cross-efficiency of a customized virtual DMU

    • Benevolent (customized)
    • Aggressive (customized)
  5. Game Cross Efficiency model: Nash Equilibrium model (Liang, et al 2008; Wu, et al 2009) (博弈交叉效率模型)
  6. Undesirable outputs, desirable inputs (非期望产出(越少越好的产出) 和 期望投入(越多越好的投入))
  7. Nondiscretionary input/output model (不可随意控制的投入产出)
  8. Preference (weighted) model (Set Input/output Weights) (偏好(加权)模型,设置投入产出权重)
  9. Restricted multiplier model (Cone Ratio DEA, Restricted Cone DEA, Assurance Region DEA) (权重比值约束模型,锥比率DEA,约束锥DEA,保证域DEA)
  10. Meta-Frontier and Non-Concave Meta-Frontier (共同前沿和非凸性共同前沿)
  11. Innovative Two-Stage and Generalized Network DEA (新型两阶段网络DEA和广义网络DEA)
What's more important, MaxDEA X provides most possible combinations of the above models.最为重要的是, MaxDEA X 提供了尽可能多的上述模型选项的组合应用。