Initial public offerings often show abnormal fist-day returns. These, usually referred to as underpricing, are difficult to predict. Among the main obstacles, we could mention challenges like the fact that not all relevant variables have been identified yet; the mix of weak and strong indicators or the prevalence of outliers. In this context, we suggest that adaptive neuro-fuzzy inference systems and fuzzy rule-based system with genetic optimization have a lot to bring to the table. We test the predictive performance of these on a sample of 866 US IPOs and we benchmark them against six fuzzy algorithms and a set of eight classic machine learning alternatives. We conclude that both fuzzy systems, especially the former should be seriously considered in this domain.
AhaD.W., KiblerD. and M.K.Albert, Instance-based learningalgorithms, Machiche Learn6 (1991), 37–66.
2.
AiF.-Y. and J.-Y.Yang, Approaches to dynamic multiple attributedecision making with 2-tuple linguistic information, J Intell Fuzzy Syst27 (2014), 2715–2723.
3.
AlbringS.M., ElderR.J. and J.Zhou, IPO underpricing and auditquality differentiation within non-big 5 firms, Int J Audit11 (2007), 115–131.
4.
AlcaláR. and F.Herrera, Genetic Tuning on Fuzzy Systems Basedon the Linguistic 2-Tuples Representation, in: Proc 2004 IEEEInt Conf Fuzzy Syst2004, 233–238.
BagheriA., PeyhaniH.M. and M.Akbari, Financial forecasting usingANFIS networks with Quantum-behaved Particle Swarm Optimization, Expert Syst Appl41 (2014), 6235–6250.
7.
BarakS., DahooieJ.H. and T.Tichý, Wrapper ANFIS-ICA method todo stock market timing and feature selection on the basis of Japanese Candlestick, Expert Syst Appl42 (2015), 9221–9235.
8.
BellR.G., FilatotchevI. and R.V.Aguilera, Corporate governanceand investors’ perceptions of foreign IPO value: An institutionalperspective, Acad Manag J57 (2014), 301–320.
9.
BenvenisteL.M. and P.A.Spindt, How investment bankers determinethe offer price and allocation of new issues, J Financ Econ24 (1989), 343–361.
10.
BrealeyR., LelandH.E. and D.H.Pyle, Informationa asymmetries,financial structure and financial intermediation, J Finance32 (1977), 371–387.
11.
CasillasJ., CordónO., HerreraF. and P.Villar, A hybridlearning process for the knowledge base of a fuzzy rule-basedsystem, in: Proc 2004 Int Conf Inf Process Manag Uncertain Knowledge-Based Syst2001, pp. 2189–2196.
12.
ChalkA.J. and J.W.P.III, Initial public offerings: Daily returns,offering types and the price effect, Financ Anal J43 (1987), 65–69.
13.
ChenY.-S., ChenJ.R.-S. and C.-H.Cheng, An alternate method ofexamining IPO returns, Intell Autom Soft Comput16 (2010), 151–162.
14.
ChengC.-H., ChenY.-S. and J.-S.Chen, Classifying initial returnsof electronic firm’s IPOs using entropy based rough sets in Taiwantrading systems, in: Second Int Conf Innov Comput Inf ControlICICIC 2007, 2008.
15.
ChengC.-H. and J.-W.Liu, OWA rough set to forecast the industrialgrowth rate, in: Int Conf Converg Inf Technol ICCIT 20072007, pp. 1862–1867.
16.
ChiuS., Fuzzy model identification based on cluster estimation, J Intell Fuzzy Syst2 (1994), 267–278.
17.
ChiuS., Method and software for extracting fuzzy classificationrules by subtractive clustering, 1996 Bienn Conf North Am Inf Process Soc, IEEE1996, pp. 461–465.
18.
ChoensawatW. and P.Polsiri, Financial institution failureprediction using adaptive neuro-fuzzy inference systems: Evidencefrom the east Asian economic crisis, J Adv Comput Intell Intell Informatics17 (2013), 83–92.
19.
CordónO., GomideF.A.C., HerreraF., HoffmannF. and MagdalenaL., Ten years of genetic fuzzy systems: Current framework andnew trends, Fuzzy Sets Syst141 (2004), 5–31.
20.
CordónO., HerreraF., MagdalenaL. and P.Villar, A geneticlearning process for the scaling factors, granularity and contextsof the fuzzy rule-based system data base, Inf Sci (Ny)136 (2001), 85–107.
21.
DoubravskýK., MeluzínT. and M.Dohnal, Decision-making onimplementation of IPO under topological uncertainty, Acta Univ Agric Silvic Mendelianae Brun63 (2015), 193–200.
22.
EsfahanipourA., GoodarziM. and R.Jahanbin, Analysis andforecasting of IPO underpricing, Neural Comput Appl27 (2016), 651–658.
23.
EshelmanL.J., The CHC adaptive search algorithm: How to have safeserach when engaging in nontraditional genetic recombination, Found Genet Algorithms1991, 265–283.
24.
FrankM.H.E., HolmesG., PfahringerB., ReutemannP.and WittenI.H., The WEKA data mining software: An update, SIGKDD Explor Newsl11 (2009), 10–18.
25.
GhaziH.L. and M.S.Abadeh, Mining fuzzy association rules with2-tuple linguistic terms in stock market data by using geneticalgorithm, in: AISP 2012 - 16th CSI Int Symp Artif Intell Signal Process2012, pp. 354–359.
26.
GuanJ., ShiD., ZuradaJ.M. and A.S.Levitan, Analyzing massivedata sets: An adaptive fuzzy neural approach for prediction, with areal estate illustration, J Organ Comput Electron Commer24 (2014), 94–112.
27.
HanleyK.W., The underpricing of initial public offerings and thepartial adjustment phenomenon, J Financ Econ34 (1993), 231–250.
28.
HerreraF., MartínezL. and F.Herrera, A 2-tuple fuzzylinguistic representation model for computing with words, IEEETrans Fuzzy Syst8 (1999), 746–752.
29.
HolmesG., HallM. and E.Prank, Generating Rule Sets from Model Trees, in: Adv Top Artif Intell1999, pp. 1–12.
30.
HuangC.-F., TsaiM.-Y., HsiehT.-N., KuoL.-M. and B.R.Chang, Astudy of hybrid genetic-fuzzy models for IPO stock selection, in: 2012 Int Conf Fuzzy Theory Its Appl iFUZZY 20122012, pp. 357–362.
31.
IshibuchiH., NozakiK., TanakaH., HosakaY. and M.Matsuda, Empirical study on learning in fuzzy systems by rice taste analysis, Fuzzy Set Syst64 (1994), 129–144.
32.
JangJ.-S.R., ANFIS: Adaptive-network-based fuzzy inference system, IEEE Trans Syst Man Cybern23 (1993), 665–685.
33.
KasabovN.K. and Q.Song, DENFIS: Dynamic evolving neural-fuzzyinference system and its application for time-series prediction, IEEE Trans Fuzzy Syst10 (2002), 144–154.
34.
KimJ. and N.Kasabov, HyFIS: Adaptive neuro-fuzzy inference systemsand their application to nonlinear dynamical systems, Neural Networks12 (1999), 1301–1319.
35.
KirkulakB. and C.Davis, Underwriter reputation and underpricing:Evidence from the Japanese IPO market, Pacific-Basin Financ J13 (2005), 451–470.
36.
LjungqvistA. and W.J.W.Jr., IPO pricing in the dot-com bubble, J Finance58 (2003), 723–752.
37.
LowryM. and K.J.Murphy, Executive stock options and IPOunderpricing, J Financ Econ85 (2007), 39–65.
38.
LowryM., OfficerM.S. and G.W.Schwert, The variability of IPOinitial returns, J Finance65 (2010), 425–465.
39.
MaitiS. and R.K.Tiwari, A comparative study of artificial neuralnetworks, Bayesian neural networks and adaptive neuro-fuzzyinference system in groundwater level prediction, Environ Earth Sci71 (2014), 3147–3160.
40.
MannH.B. and D.R.Whitney, On a test of whether one of two randomvariables is stochastically larger than the other, Ann Math Stat18 (1947), 50–60.
41.
MegginsonW.L. and K.A.Weiss, Venture capitalist certification ininitial public offerings, J Finance46 (1991), 879–903.
42.
MengD. and Y.Xu, Credit rating method with heterogeneous information, Lect Notes Comput Sci (Including Subser Lect Notes Artif Intell LectNotes Bioinformatics) 5009 LNAI2008, pp. 739–746.
43.
MoodyJ. and C.J.Darken, Fast learning in networks of locally-tunedprocessing units, Neural Comput1 (1989), 281–294.
44.
NomuraH., HayashiI. and N.Wakami, A learning method of fuzzyinference rules by descent method, in: Proc 1992 IEEE Int Conf Fuzzy Syst1992, pp. 203–210.
45.
PedryczW., Associations and rules in data mining: A link analysis, Int J Intell Syst19 (2004), 653–670.
46.
PetkovicD., Adaptive neuro-fuzzy approach for estimation of windspeed distribution, Int J Electr Power Energy Syst73 (2015), 389–392.
47.
QuinlanJ.R.Learning with continuous classes, in: Proc Aust Jt Conf Artif Intell, Singaore1992, 343–348.
48.
RitterJ.R., Equilibrium in the initial public offerings market, Annu Rev Financ Econ3 (2011), 347–374.
49.
RitterJ.R. and I.Welch, A review of IPO activity, pricing andallocations, J Finance57 (2002), 1795–1828.
50.
RodgersJ.L. and A.W.Nicewander, Thirteen ways to look at thecorrelation coefficient, Am Stat42 (1988), 59–66.
51.
RousseeuwP.J., LeroyA.M.Robust Regression and Outlier Detection, John Wiley & Sons, Inc, 1987.
52.
RumelhartD.E., HintonG.E. and R.J.Williams, Learning Representations by Back-propagating Errors, Nature323 (1986), 533–536.
53.
SmartS.B. and C.J.Zutter, Control as a motivation forunderpricing: A comon of dual and single-class IPOs, JFinanc Econ69 (2003), 85–110.
54.
SmolaA.J. and B.Schölkopf, A tutorial on support vectorregression, Stat Comput14 (2004), 199–222.
55.
TengL., YangL.X., TangF., TengS. and W.Zhang, 3-dimensionevaluation method for stock investment based on 2-tuple linguistic, Lect Notes Comput Sci (Including Subser Lect Notes Artif Intell Lect Notes Bioinformatics)8944 (2015), 326–339.
56.
TengY.W. and W.J.Wang, Constructing a user-friendly GA-based fuzzysystem directly from numerical data, IEEE Trans Syst Man,Cybern Part B34 (2004), 2060–2070.
57.
VellaV. and W.L.Ng, Enhancing risk-adjusted performance of stockmarket intraday trading with Neuro-Fuzzy systems, Neurocomputing141 (2014), 170–187.
58.
WangL.X. and J.M.Mendel, Generating fuzzy rules by learning fromexamples, IEEE Trans Syst Man Cybern22 (1992), 1414–1427.
59.
WeiL.-Y., A GA-weighted ANFIS model based on multiple stock marketvolatility causality for TAIEX forecasting, Appl Soft Comput J13 (2013), 911–920.
60.
XiaohuiG., RitterJ.R. and Z.Zhu, Where have all the IPOs gone?J Financ Quant Anal48 (2013), 1663–1692.
61.
YagerR.R. and D.P.Filev, Generation of fuzzy rules by mountainclustering, J Intell Fuzzy Syst2 (1994), 209–219.
