Browse > Article
http://dx.doi.org/10.12989/was.2018.26.6.383

Alternative robust estimation methods for parameters of Gumbel distribution: an application to wind speed data with outliers  

Aydin, Demet (Department of Statistics, Faculty of Science and Letters, Sinop University)
Publication Information
Wind and Structures / v.26, no.6, 2018 , pp. 383-395 More about this Journal
Abstract
An accurate determination of wind speed distribution is the basis for an evaluation of the wind energy potential required to design a wind turbine, so it is important to estimate unknown parameters of wind speed distribution. In this paper, Gumbel distribution is used in modelling wind speed data, and alternative robust estimation methods to estimate its parameters are considered. The methodologies used to obtain the estimators of the parameters are least absolute deviation, weighted least absolute deviation, median/MAD and least median of squares. The performances of the estimators are compared with traditional estimation methods (i.e., maximum likelihood and least squares) according to bias, mean square deviation and total mean square deviation criteria using a Monte-Carlo simulation study for the data with and without outliers. The simulation results show that least median of squares and median/MAD estimators are more efficient than others for data with outliers in many cases. However, median/MAD estimator is not consistent for location parameter of Gumbel distribution in all cases. In real data application, it is firstly demonstrated that Gumbel distribution fits the daily mean wind speed data well and is also better one to model the data than Weibull distribution with respect to the root mean square error and coefficient of determination criteria. Next, the wind data modified by outliers is analysed to show the performance of the proposed estimators by using numerical and graphical methods.
Keywords
Gumbel distribution; least absolute deviation; median/MAD estimator; Monte-Carlo simulation; wind speed data;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Akgul, F.G., Senoglu, B. and Arslan, T. (2016), "An alternative distribution to Weibull for modeling the wind speed data: Inverse Weibull distribution", Energ. Convers. Manage., 114, 234-240.   DOI
2 Apery, R. (1979), "Irrationalite de ${\zeta}$ (2) et ${\zeta}$ (3)", Asterisque, 61, 11-13.
3 Arslan, T., Acitas, S. and Senoglu, B. (2017), "Generalized Lindley and power Lindley distributions for modeling the wind speed data", Energ. Convers. Manage., 152, 300-311.   DOI
4 Aydin, D. and Senoglu, B. (2015), "Monte Carlo comparison of the parameter estimation methods for the two-parameter Gumbel distribution", J. Modern Appl. Stat. Method., 14(2), 123-140.   DOI
5 Aydin, D. (2018), "Estimation of the lower and upper quantiles of Gumbel distribution: an application to wind speed data", Appl. Ecology Environ. Res., 16(1), 1-15.   DOI
6 Barnett, V. and Lewis, T. (1994), "Outliers in statistical data", John Wiley, New York, USA.
7 Bernard, A. and Bosi-Levenbach, E.C. (1953), "The plotting of observations on probability paper", Statistica Neerlandica, 7, 163-173.   DOI
8 Blom, G. (1958), Statistical estimates and transformed beta variables, John Wiley and Sons, New York, USA.
9 Boudt, K., Caliskan, D. and Croux, C. (2011), "Robust explicit estimators of Weibull parameters", Metrika, 73, 187-209.   DOI
10 Croux, C., Filzmoser, P., Pison, G. and Rousseeuw, P.J. (2003), "Fitting multiplicative models by robust alternating regressions", Stat. Comput., 13, 23-36.   DOI
11 Corsini, G., Gini, F., Greco, M.V. and Verrazzani, L. (1995), "Cramer-Rao bounds and estimation of the parameters of the Gumbel distribution", Aerosp. Electron. Syst., 31(3), 1202-1204.   DOI
12 Cox, N.J. (2005), "Speaking Stata: density probability plots", Stata J., 5(2), 259-273.
13 Davis, R.A. and Dunsmuir, W.T.M. (1997), "Least absolute deviation estimation for regression with ARMA errors", J. Theor. Probab., 10, 481-497.   DOI
14 Dixon, W.J. (1950), "Analysis of extreme values", Annal. Math. Stat., 21, 488-506.   DOI
15 Dunsmuir, W.T.M. and Spencer, N.M. (1991), "Strong consistency and asymptotic normality of L1 estimates of the autoregressive moving-average model", J. Time Series Anal., 12, 95-104.   DOI
16 Bai, J. (1995), "Least absolute deviation estimation of a shift", Econometric Theory, 11(3), 403-436.   DOI
17 Ercelebi, S.G. and Toros, H. (2009), "Extreme value analysis of Istanbul air pollution data", Clean Soil Air Water, 37, 122-131.   DOI
18 Gumbel, E.J. (1941), "The return period of flood flows", Annal. Math. Stat., 12(2), 163-190.   DOI
19 Fischer, S., Fried, R. and Schumann, A. (2015), "Examination for robustness of parametric estimators for flood statistics in the context of extraordinary extreme events", Hydrology Earth Syst. Sci., 12, 8553-8576.   DOI
20 Graybeal, D. and Leathers, D. (2006), "Snowmelt-related flood risk in Appalachia: first estimates from a historical snow climatology", J. Appl. Meteorol. Clim., 45(1), 178-193.   DOI
21 Herd, G.R. (1960), "Estimation of reliability from incomplete data", Proceedings of the 6th National Symposium on Reliability and Quality Control, IEEE, New York.
22 Hong, H.P., Li, S.H. and Mara, T.G. (2013), "Performance of the generalized least-squares method for the Gumbel distribution and its application to annual maximum wind speeds", J. Wind Eng. Ind. Aerod., 119, 121-132.   DOI
23 Jenkinson, A.F. (1955), "The frequency distribution of the annual maximum (or minimum) values of meteorological elements", Q. J. Roy. Meteor. Soc., 81, 158-171.   DOI
24 Johnson, L.G. (1964), The statistical treatment of fatigue experiments, Elsevier, New York, USA.
25 Kang, D., Ko, K. and Huh, J. (2015), "Determination of extreme wind values using the Gumbel distribution", Energy, 86, 51-58.   DOI
26 Kantar, Y.M. and Yildirim, V. (2015), "Robust estimation for parameters of the extended Burr type III distribution", Commun. Statistics-Simulation and Comput., 44, 1901-1930.   DOI
27 Kantar, Y.M.; Usta, I.; Arik, I. and Yenilmez, I. (2018), "Wind speed analysis using the Extended Generalized Lindley Distribution", Renew. Energy, 118, 1024-1030.   DOI
28 Kollu, R., Rayapudi, S.R., Narasimham, S.V.L. and Pakkurthi, K.M. (2012), "Mixture probability distribution functions to model wind speed distributions", Int. J. Energ. Environ. Eng., 3, 1-10.   DOI
29 Kao, J.H. (1958), "Computer methods for estimating Weibull parameters in reliability studies", IRE T. Reliab. Quality Control, 13, 15-22.
30 Kao, J.H. (1959), "A graphical estimation of mixed Weibull parameters in life-testing of electron tubes", Technometrics, 1(4), 389-407.   DOI
31 Koutsoyiannis, D. (2004), "Statistics of extremes and estimation of extreme rainfall: I. theoretical investigation", Hydrol. Sci. J., 49(4), 575-590.   DOI
32 Landwehr, J.M., Matalas, N.C. and Wallis, J.R. (1979), "Probability weighted moments compared with some traditional techniques in estimating Gumbel parameters and quantiles", Water Resour. Res., 15(5), 1055-1064.   DOI
33 Lee, B.H., Ahn, D.J., Kim, H.G. and Ha, Y.C. (2012), "An estimation of the extreme wind speed using the Korea wind map", Renew Energ., 42, 4-10.   DOI
34 Ling, S. (2005), "Self-weighted LAD estimation for infinite variance autoregressive models", J. Roy. Stat. Soc.: Series B, 67, 381-393.   DOI
35 Mohammadi, K., Alavi, O. and McGowan, J.G. (2017), "Use of Birnbaum-Saunders distribution for estimating wind speed and wind power probability distributions: A review", Energ. Convers. Manage., 143, 109-122   DOI
36 Morgan, E.C., Lackner, M., Vogel, R.M. and Baise, L.G. (2011), "Probability distribution for offshore wind speeds", Energ. Convers. Manage., 52, 15-26.   DOI
37 Mousa, M.A.M., Jaheen, Z.F. and Ahmad, A.A. (2002), "Bayesian estimation, prediction and characterization for the Gumbel model based on records", J. Theor. Appl. Stat., 36(1), 65-74.
38 Pishgar-Komleh, S.H., Keyhani, A. and Sefeedpari, P. (2015), "Wind speed and power density analysis based on Weibull and Rayleigh distributions (a case study: Firouzkooh county of Iran)", Renew. Sust. Energ. Rev., 42, 313-322.   DOI
39 Olive, D. (2006), "Robust estimators for transformed location scale families", http://lagrange.math.siu.edu/Olive/pprloc3.pdf.
40 Pan, J., Wang, H. and Yao, Q. (2007), "Weighted least absolute deviations estimation for ARMA models with infinite variance", Econometric Theory, 23(5), 852-879.   DOI
41 Pollard, D. (1991), "Asymptotics for least absolute deviation regression estimators", Econometric Theory, 7, 186-199.   DOI
42 Phien, H.N. (1987), "A review of methods of parameter estimation for the extreme-value type I distribution", J. Hydrol., 86, 391-398.
43 Rasmussen, P.F. and Gautam, N. (2003), "Alternative PWMestimators of the Gumbel distribution", J. Hydrol., 280, 265-271.   DOI
44 Raynal, J.A. and Salas, J.D. (1986), "Estimation procedures for the type-1 extreme value distribution", J. Hydrol., 87, 315-336.   DOI
45 Rousseeuw, P.J. (1984), "Least median of squares regression", J. Am. Stat. Association, 79, 871-880.   DOI
46 Rousseeuw, P.J. and Leroy, A. (1987), Robust regression and outlier detection, Wiley, New York, USA.
47 Simiua, E.I., Heckertb, N.A., Filliben, J.J. and Johnson, S.K. (2001), "Extreme wind load estimates based on Gumbel distribution of dynamic pressures: an assessment", Struct. Saf., 23, 221-229.   DOI
48 Trzpiot, G. (2009), "Extreme value distributions and robust estimation", Folia Oeconomica, 228, 85-92.
49 Sohoni, V., Gupta, S. and Nema, R. (2016), "A comparative analysis of wind speed probability distributions for wind power assessment of four sites", Turkish J. Elec. Eng. Comput. Sci., 24, 4724-4735.   DOI
50 Soukissian, T. (2013) "Use of multi-parameter distributions for offshore wind speed modeling: The Johnson SB distribution", Appl. Energ., 111, 982-1000.   DOI
51 Wolstenholme, L.C. (1999), Reliability modelling: a statistical approach, Chacman and Hall Press, New York, USA.
52 Yavuz, A.A. (2013), "Estimation of the shape parameter of the Weibull distribution using linear regression methods: noncensored samples", Qual. Reliab. Eng. Int., 29, 1207-1219.   DOI
53 Zyl, J.M.V. and Schall, R. (2012), "Parameter estimation through weighted least-squares rank regression with specific reference to the Weibull and Gumbel distributions", Commun. Stat. - Simul. C., 41, 1654-1666.   DOI
54 Tiryakioglu, M. and Hudak, D. (2007), "On estimating Weibull modulus by the linear regression method", J. Mater. Sci., 42, 10173-10179.   DOI