Abstract
The iteration algorithm of computation of effective estimators of the shape parameters of beta distributions using the unbiased estimators of the end point parameters of the random variable were obtained and investigated. For the cases when more accurate estimations of the parameters are required, one more step of computation, realized optimization of the obtained estimations, is necessary. The computation results, realized on the basis of the simulation of the appropriate random samples, demonstrate the correctness of the obtained theoretical outcomes.
Similar content being viewed by others
References
AbouRizk, S.M., Halpin, D.W. and Wilson, J.R. (1991). Visual iterative fitting of beta distribution. J. Constr. Eng. Manag.117, 589–605.
AbouRizk, S.M., Halpin, D.W. and Wilson, J.R. (1994). Fitting beta distributions based on sample data. J. Constr. Eng. Manag.120, 2, 288–305.
Bazaraa, M.S. and Shetty, C.M. (1979). Nonlinear Programming Theory and Algorithms. Wiley, New York.
Boll, R.M., Connell, J.H., Pankanti, S.H., Ratha, N.K. and Senior, A.W. (2004). Guide to Biometrics. Springer, New York.
David, H.A. and Nagaraja, H.N. (2003). Order Statistics, 3rd edn. Wiley Series in Probability and Statistics, New York, Inc.
Demetrashvili, N. and Van den Heuvel, E.R. (2015). Confidence intervals for intraclass correlation coefficients in a nonlinear dose-response meta-analysis. Biometrics71, (2), 548–555.
Dishon, M. and Weiss, G.H. (1980). Small sample comparison of estimation methods for the beta distribution. J. Stat. Comput. Simul.11, 1–11.
Hillier, F.S. and Lieberman, G.J. (1980). Introduction to Operation Research, 3rd edn. Holden-Day, San Francisco.
Jonson, N.L., Kotz, S. and Balakrishnan, N. (2004). Continuous Univariate Distributions, vol. 2, 2nd edn. Wiley, New York.
Kachiashvili, K.J. and Prangishvili, A.I. (2016). Verification in biometric systems: problems and modern methods of their solution. J. Appl. Stat., 1–20, https://doi.org/10.1080/02664763.2016.1267122.
Kachiashvili, K.J. and Stepanishvili, V.A. (1988). Estimators of unknown parameters of some non-regularities probability distribution densities. Avtometria2, 109–111.
Kachiashvili, K.J. and Topchishvili, A.L. (2016). Estimators of the parameters of irregular Right-Angled triangular distribution. Model. Assist. Stat. Appl.11, 179–184.
Pastushko, O.N. and Nevludov, I. (2012). Analysis of quality indicators of biometrical systems of authentication of users. Problems of Telecommunication, 4. http://pt.journal.kh.ua.
Primak, A.V., Kafarov, V.V. and Kachiashvili, K.J. (1991). System Analysis of Control and Management of Air and Water Quality. Naukova Dumka, Kiev. (Science and Technical Progress).
Acknowledgments
We thank the editor and unknown reviewers for their help and attention to our work.
Funding
This research was supported by Shota Rustaveli National Science Foundation of Georgia grant AR/183/4-100/13.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1. Formulae for Determination of Direct and Inverse Incomplete Beta-Functions
Let us consider a standard form of the incomplete beta distribution, i.e.,
B(p,q) = B1(p,q) is a complete beta function and
Let us introduce denotations:
Then we can represent the beta function in the form of the series
or
where
At integer values of the parameters, the following relations are correct
In particular,
At half-integer values of the parameters, we have
where
In particular,
For computation of the inverse incomplete beta function, let us introduce Q(α,β,ξ), which is the inverse function of Iz(α,β) for given α and β. It satisfies the relation
At ξ → 0
where
At ξ → 1
where
For determination of the inverse incomplete beta function, we used the following iteration method.
Let us consider the equation
at 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, and (1 − p)(1 − q)≠ 0. Let us denote
For the solution of (1), the method of Newton’s iteration was used. The initial approximation x0 of the iterative process is to be defined as follows:
if p > 1 and q > 1 then x0 = xμ;
if p ≤ 1 and q ≥ 1 then x0 = 0;
if p ≥ 1 and q ≤ 1 then x0 = 1;
if α < 1 and β < 1 and \(I_{x_{\mu }}(p,q) >y\) then x0 = 0;
if α < 1 and β < 1 and \(I_{x_{\mu }}(p,q) <y\) then x0 = 1.
It should be noted that the monotonous convergence of the iteratively obtained sequence of the solutions of (1) to the desired value is guaranteed.
Appendix 2. Examples
Examples of the graphs of the divergences between empirical and estimated distribution functions by the samples of different size n for case 11 at a = 12.0, b = 21.0, p = 7.21; q = 1.252 (see Table 6) are represented in Fig. 6.
Rights and permissions
About this article
Cite this article
Kachiashvili, K.J., Melikdzhanjan, D.I. Estimators of the Parameters of Beta Distribution. Sankhya B 81, 350–373 (2019). https://doi.org/10.1007/s13571-018-0157-2
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13571-018-0157-2
Keywords and phrases
- Beta distribution
- Maximum likelihood estimator
- Biased estimator
- Unbiased estimator
- Iteration algorithm
- Optimization algorithm