Abstract
With reference to pairwise comparisons in group decision-making, consider the problem of how a decision-maker, who knows his/her own assessment and the aggregate assessment of all decision-makers, can predict the assessments of the other decision-makers. We address this problem from a statistical perspective and propose statistical prediction intervals. These are seen to be often significantly shorter and hence appreciably more powerful than their deterministic counterparts available in the literature, accounting for the confidence level as well. In the process, analytical closed form solutions for these deterministic intervals are also obtained. Extensive computations and simulations show that our statistical approach remains very robust to departure from underlying assumptions, frequently being even more efficient. Moreover, this approach is found to be useful generally in predicting an unknown individual component of a given total, thus having much wider applicability beyond the immediate context of pairwise comparisons in group decision-making.
Similar content being viewed by others
References
Aczel J, Alsina C (1986) On synthesis of judgements. Socio-Econ Plan Sci 20:333–339
Aczel J, Saaty TL (1983) Procedures for synthesizing ratio judgments. J Math Psychol 27:93–102
Al-Omush A, Al Shishany A, Sarhan NM, Alakaleek W (2020) Enhancing anonymity in group decision supported systems meetings: participants’ physical proximity challenge. J Manage Inform Decis Sci 23:491–498
Brunelli M (2019) A study on the anonymity of pairwise comparisons in group decision making. Eur J Oper Res 279:502–510
Chao X, Kou G, Peng Y, Viedma EH (2021) Large-scale group decision-making with non-cooperative behaviors and heterogeneous preferences: an application in financial inclusion. Eur J Oper Res 288:271–293
Hummel JM, Bridges JFP, IJzerman MJ (2014) Group decision making with the analytic hierarchy process in benefit-risk assessment: a tutorial. Patient 7:129–140
Ishizaka A, Labib A (2011) Review of the main developments in the analytic hierarchy process. Expert Sys Appl 38:14336–14345
Jalao ER, Wu T, Shunk D (2014) A stochastic AHP decision making methodology for imprecise preferences. Inform Sci 270:192–203
Jessup LM, Connolly T, Galegher J (1990) The effects of anonymity on GDSS group process with an idea-generating task. MIS Quart 14:313–321
Keeney RL, Raiffa H (1993) Decisions with multiple objectives: preferences and value tradeoffs, 2nd edn. Cambridge University Press, Cambridge
Kou G, Ergu D, Lin C, Chen Y (2016) Pairwise comparison matrix in multiple criteria decision making. Technol Econ Dev Eco 22:738–765
Kulakowski K, Mazurek J, Strada M (2020) On the similarity between ranking vectors in the pairwise comparison method. arXiv: 2010.04778v1
Lehmann EL (1986) Testing statistical hypotheses, 2nd edn. Springer, New York
Li KW, Wang ZJ, Tong X (2016) Acceptability analysis and priority weight elicitation for interval multiplicative comparison matrices. Eur J Oper Res 250:628–638
Lin C, Kou G, Ergu D (2014) A statistical approach to measure the consistency level of the pairwise comparison matrix. J Oper Res Soc 65:1380–1386
Lin C, Kou G (2015) Bayesian revision of the individual pair-wise comparison matrices under consensus in AHP–GDM. Appl Soft Comput 35:802–811
Liu S, Yu W, Chan FTS, Niu B (2020) A variable weight-based hybrid approach for multi-attribute group decision making under interval-valued intuitionistic fuzzy sets. Int J Intell Syst. https://doi.org/10.1002/int.22329
Matsui M (2017) Prediction of components in random sums. Methodol Comput Appl Probab 19:573–587
Nunamaker JF Jr, Deokar AV (2008) GDSS parameters and benefits. In: Burstein F, Holsapple CY (eds) Handbook on decision support systems 1: basic themes. Springer, Berlin, pp 391–414
Olds EG (1952) A note on the convolution of uniform distributions. Ann Math Stat 23:282–285
Perez-Rodriguez F, Rojo-Alboreca A (2012) Forestry application of the AHP by use of MPC© software. For Syst 21:418–425
Rauch W (1979) The decision Delphi. Technol Forecast Soc Change 15:159–169
Saaty TL (1977) A scaling method for priorities in hierarchical structures. J Math Psychol 15:234–281
Valacich JS, Jessup LM, Dennis AR, Nunamaker JF Jr (1992) A conceptual framework of anonymity in group support systems. Group Decis Negot 1:219–241
Wu W, Kou G, Peng Y, Ergu D (2012) Improved AHP-group decision making for investment strategy selection. Technol Econ Dev Eco 18:299–316
Acknowledgements
We thank the referee, associate editor and editor for very constructive suggestions. The work of Rahul Mukerjee was supported by the J.C. Bose National Fellowship of the Government of India and a grant from the Indian Institute of Management Calcutta.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: Proofs
Appendix: Proofs
Proof of Proposition 1. First, note that bm ≥ \(b_{m}^{ - }\), because by (2) and (3), bm ≥ L and bm = \((b - \Sigma_{k = 1}^{m - 1} \lambda_{k} b_{k} )/\lambda_{m}\) ≥ \((b - \lambda_{1} b_{1} - U\Sigma_{k = 2}^{m - 1} \lambda_{k} )/\lambda_{m}\) = \(\tilde{L}\). To see the existence of a feasible solution attaining this lower bound, first let L > \(\tilde{L}\). Then \(b_{m}^{ - }\) = L, which is attained when b2 = … = bm–1 = \((b - \lambda_{1} b_{1} - \lambda_{m} L)/\Sigma_{k = 2}^{m - 1} \lambda_{k}\), bm = L. This solution is feasible because it obviously meets (3); moreover, it meets (2) as
by (4), while
Next, if \(\tilde{L}\) ≥ L, then \(b_{m}^{ - }\) = \(\tilde{L}\), which is attained when b2 = … = bm–1 = U, bm = \(\tilde{L}\). This is again a feasible solution because it obviously meets (3); it also satisfies (2) as \(\tilde{L}\) ≥ L, and by (4),
Thus, the claim about \(b_{m}^{ - }\) is proved. The claim about \(b_{m}^{ + }\) follows similarly.
Proof of Theorem 1
Let \(\tilde{S}\) = \(\Sigma_{k = 1}^{n - 1} Y_{k}\). Following Olds (1952), with a more compact notation, the pdfs of \(\tilde{S}\) and S are given, respectively, by
Because Yn and \(\tilde{S}\) are independent and Yn is uniform on [0, cn], their joint pdf equals
Since \(\tilde{S}\) = S − Yn, the transformation (Yn,\(\tilde{S}\)) → (Yn, S) yields the joint pdf of Yn and S as
recalling (7). Dividing this by the pdf of S as stated above, the result follows.
Proof of Theorem 2
We first present the following lemma.
Lemma A1
Let n ≥ 2. Then (a) \(g_{n} (s)\) = \(g_{n} (n - s)\), 0 ≤ s ≤ n, (b) \(g_{n} (s)\) is increasing in s on [0, n/2] and decreasing in s on [n/2, n].
Proof. (a) This is known in the literature on Irwin–Hall distribution that concerns the sum of independent random variables, each uniform on [0, 1], and can be deduced from the fact that
(b) This also seems to be known as a folklore. A formal proof is presented here as we could not find one in the literature. By (18), (b) holds for n = 2, for then \(g_{2} (s)\) equals s if 0 ≤ s ≤ 1, and 2 − s if 1 ≤ s ≤ 2. To apply induction, suppose (b) holds for n = p (≥ 2), and consider n = p + 1. By (18), \(g_{p + 1} (s)\) is differentiable, with derivative
using (18) and the fact that
By (18), \(g_{p} (s - 1)\) = 0 if \(s - 1\) ≤ 0. Hence, if 0 < s ≤ p/2, then by (A.1) and induction hypothesis, \(g^{\prime}_{p + 1} (s)\) > 0. Moreover, if p/2 < s < (p + 1)/2, then s − 1 < p − s < p/2, so that by (A.1), part (a) and induction hypothesis,
Thus, \(g^{\prime}_{p + 1} (s)\) > 0 for 0 < s < (p + 1)/2, i.e., \(g_{p + 1} (s)\) is increasing in s on [0, (p + 1)/2]. Hence, by part (a), \(g_{p + 1} (s)\) is decreasing in s on [(p + 1)/2, p + 1], and the result follows by induction.
Proof of Theorem 2.
(I) Clearly, s − yn decreases from s − \(Y_{n}^{ - }\) to s − \(Y_{n}^{ + }\) as yn increases from \(Y_{n}^{ - }\) to \(Y_{n}^{ + }\). From (16), observe that the following hold for n ≥ 3:
-
(a)
If 0 < s ≤ (n − 1)/2, then \(Y_{n}^{ - }\) = 0, s − \(Y_{n}^{ - }\) = s ≤ (n − 1)/2 and s − \(Y_{n}^{ + }\) = max(s − 1, 0) ≥ 0.
-
(b)
If (n + 1)/2 ≤ s < n, then \(Y_{n}^{ + }\) = 1, s − \(Y_{n}^{ - }\) = min(s, n − 1) ≤ n − 1, and s − \(Y_{n}^{ + }\) = s − 1 ≥ (n − 1)/2.
-
(c)
If (n − 1)/2 < s < (n + 1)/2, then \(Y_{n}^{ - }\) = 0, \(Y_{n}^{ + }\) = 1, s − \(Y_{n}^{ - }\) = s > (n − 1)/2, s − \(Y_{n}^{ + }\) = s − 1 < (n − 1)/2, and s − yn equals (n − 1)/2 when yn equals s − (n − 1)/2 = s*.
The result now follows if one invokes Lemma A1(b) with n there replaced by n − 1, and recalls (17).
(II) This is intuitively expected from (I) but a formal constructive proof will help. For ease in notation, write y for yn and h(y) for \(h(y_{n} |s)\), the cumulative distribution function associated with h(y) being denoted by H(y). As h(y) is continuous and hence bounded on the compact interval D, it follows that H(y) is continuous on D. So, there exist d1 and d2 in the interior of D such that H(d1) = \(\alpha\) and H(d2) = \(1 - \alpha\). Under (a) or (b) of (I), take q(s) = h(d2) or q(s) = h(d1), for then h(y) ≥ q(s) if and only if y ≤ d2 or y ≥ d1, respectively.
Consider now (c) of (I). Then D = [0, 1]. Let \(h_{1}^{ - 1} (.)\) and \(h_{2}^{ - 1} (.)\) be the piecewise inverses of h(.) on [0, s*] and [s*, 1], respectively. These inverse functions are well defined and continuous on their respective domains, as h(.) is continuous and strictly monotone on [0, s*] and [s*, 1]. Let hmin = min{h(0), h(1)} and hmax = h(s*). For hmin ≤ q ≤ hmax, define
and for any y \(\in\) D, by (c) of (I), note that h(y) ≥ q if and only if \(y^{(1)} (q)\) ≤ y ≤ \(y^{(2)} (q)\), so that the integral of h(y) over {y: h(y) ≥ q} equals \(H(y^{(2)} (q)) - H(y^{(1)} (q))\) = ψ(q), say. Now, by (A.2), ψ(hmin) = 1 and ψ(hmax) = 0. Since ψ(q) is continuous in q on [hmin, hmax] due to the continuity of \(h_{1}^{ - 1} (.)\), \(h_{2}^{ - 1} (.)\) and H(.), it follows that there exists q(s) in (hmin, hmax) such that ψ(q(s)) = \(1 - \alpha\), thus completing the proof.
Rights and permissions
About this article
Cite this article
Bose, M., Mukerjee, R. Shorter prediction intervals for anonymous individual assessments in group decision-making via pairwise comparisons. TOP 29, 833–857 (2021). https://doi.org/10.1007/s11750-021-00597-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11750-021-00597-y