Printer Friendly

A mixture of generalized Tukey's g distributions.

1. Introduction

The main focus of interest in financial economics is the distribution of stock market returns. Mandelbrot [1] suggested the family of stable Paretian distributions for stock market returns. Fama [2] established that the normality assumption of the empirical data does not hold as the distribution is fat tailed. Kon [3] and Tse [4] used a mixture of normal distributions for stockreturn. Fielitz and Rozelle [5] proposed a mixture of nonnormal stable distributions for stock price. Consequently, greater emphasis has been placed on using distributions which have asymmetry and leptokurtic properties. Recently Jimenez et al. [6] proposed option pricing based mixture of log-skew-normal distributions. If extreme events tend to occur more frequently than normal events, then skewness and kurtosis of nonnormal distributions play an essential role for the volatility smile.

The most important and useful characteristic of Tukey's g-h family of distributions introduced by Tukey [7] is that it covers most of the Pearsonian family of distributions. It can also generate several known distributions, for example, lognormal, Cauchy, exponential, and Chi-squared (see Martinez and Iglewicz [8], page 363). From Tukey's g-h family of distribution, we obtain g distribution, which is closely related to lognormal distribution and possesses similar properties of moments. Tukey's g-h family of distributions have been used to study financial markets. Badrinath and Chatterjee [9,10] and Mills [11] used g-h to model the return on a stock index, as well as the return on shares in several markets. Dutta and Babbel [12] found that the skewness and leptokurtic behavior of LIBOR were modeled effectively using g-h distribution. Dutta and Babbel [13] used g-h to model interest rates and options on interest rates, while Tang and Wu [14] proposed a new method for the Decomposition of Portfolio VaR. Dutta and Perry [15] and recently Jimenez and Arunachalam [16] used g-h distribution to study the operational risk for heavy tailed severity models. Jimenez and Arunachalam [17] provided explicit expressions for VaR and CVaR calculations using the family of Tukey's g-h distributions. Currently Jimenez et al. [18] studied generalization of Tukey's g-h family of distributions, when the standard normal random variable is replaced by a continuous random variable U with mean 0 and variance 1.

The subfamily of g distributions exhibits skewness and has great importance in the study of asymmetric distributions for analyzing data. This kind of distribution allows us to obtain scaled Log-symmetric distributions. Vitiello and Poon [19] considered a simple mixture of two g distributions for option pricing data. The purpose of this paper is to present a mixture of Tukey's g distributions and derive some statistical properties including the pdf and moment generating function and its properties.

The paper is organized as follows: Section 2 presents Tukey's g-h family of generalized distributions and its pdf, as well as the cumulative distribution function (cdf). In Section 3, some theoretical results of the mixture of two Tukey's g families of generalized distributions are presented and Section 4 explains the methodology of calculating estimation of parameters by the method of moments. Section 5 discusses the adjustment methodology of our proposed model to real data of Heating-Degree-Days (HDD) indices and finally, in Section 6, we conclude.

2. Tukey's g Family of Generalized Distributions

Tukey [7] introduced the family g-h distributions by means of two nonlinear transformations given by

Y = [T.sub.g,h] (Z) = 1/g(exp {gZ}-1) exp {h[Z.sup.2]/2} (1)

with g [not equal to] 0, h [member of] R, where the distribution of Z is standard normal. When these transformations are applied to a continuous random variable U with mean 0 and variance 1 such that its pdf [f.sub.U]* is symmetric about the origin and cdf [F.sub.U]*, the transformation [T.sub.g,h](U) is obtained, which henceforth will be termed Tukey's g-h generalized distribution. If h = 0, Tukey's g-h generalized distribution reduces to

[T.sub.g,0] (U) = 1/g (exp {gU}-l) (2)

which is known as Tukey's g generalized distribution.

In order to model an arbitrary random variable X using the transformation given in (2), Hoaglin and Peters [20] introduced two new parameters, A (location) and B (scale), and proposed the following linear transformation:

X = A + BY with Y = [T.sub.g,0] (U). (3)

The following properties for pdf, cdf, and quantile functions of Tukey's g generalized distribution were established by Jimenez et al. [18] in terms of the pdf and cdf of X as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (4)

where [lambda] = ln(B/[absolute value of (g)]) and [theta] = A - B/g. We say that the random variable X has a Log-symmetric distribution (such distributions are all asymmetric; see for reference Johnson et al. [21] and Stuart and Ord [22]) with three parameters : threshold ([theta]), scale (A), and shape (g), denoted by X ~ LS([lambda], g,[theta]).

The first expression of (4) allows us to obtain the following pdf associated with Tukey's g distribution. Table 1 shows the parameters of the pdf of X that we obtain using a selected set of well known symmetrical distributions (from Jimenez et al. [18]).

The nth moment of the random variable Y = [T.sub.g,0](U) is given by

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (5)

where [??] = (n - k)g and [M.sub.U](t) is the moment generating function of the random variable U, which are even function; that is, [M.sub.U](t) = [M.sub.U](-t). Table 2 shows parameters of the pdf and the moment generating function for a random variable U, using a selected set of well known symmetrical distributions.

Expression (5) allows us to obtain the moments of Tukey's g generalized distribution. The nth moment of the random variable X given by (3) can be obtained using the formula

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (6)

where B/g = sgn(g)[e.sup.[lambda]]. Note that the above expression of the nth moment does not depend on the parameter [theta]. Formulas for calculating the standardized skewness, [[beta].sub.1](X), and standardized excess kurtosis, [[beta].sub.2](X), are given by

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (7)

where sgn(.) denote the signum function. Note that these expressions only depend on the parameter g and its sign, respectively. Any LS distribution should satisfy the following test given in Stuart and Ord [22]:

[[beta].sub.2](X) - [[beta].sup.2.sub.1](X)-1 [greater than or equal to] 0. (8)

3. The Mixture of Two g Distributions

We assume that Y follows a Log-Symmetric Mixture (LSMIX) distribution. Let us assume that [f.sub.Y](y) is the weighted sum of m-component LS densities; that is,

[f.sub.Y] (y; [LAMBDA]) = [m.summation over (j=1)][[omega].sub.j][f.sub.U](y; [[lambda].sub.j], [g.sub.j], [[theta].sub.j]). (9)

We use the notation Y ~ LSMIX([LAMBDA]), where A = ([[xi].sub.1], ..., [[xi].sub.m]), and each element = ([[xi].sub.j], [[lambda].sub.j], [g.sub.j], [[theta].sub.j]) is the parameter vector that defines the jth component and probability weights, [[omega].sub.j], satisfying the conditions

[m.summation over (j=1)] [[omega].sub.j] = 1, 0 < [[omega].sub.j] < 1, for each j. (10)

where [[omega].sub.2] = 1 - [[omega].sub.1]. Begin with the fact that the quartile function is the inverse of the cdf. Thus, replacing [x.sub.q] > [[theta].sub.2] in (14), we obtain

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (15)

According to Titterington et al. [23] the two-component mixture of known distributions is set by two weights. Let

X = A +BY with Y ~ LSMIX ([LAMBDA]). (11)

Then we can assume that [f.sub.X](x) is the weighted sum of two Tukey's g mixture densities such that [g.sub.1][g.sub.2] > 0. Thus

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (12)

where, without loss of generality, we let [[theta].sub.1] < [[theta].sub.2], 0 [less than or equal to] [[omega].sub.1] [less than or equal to] 1 and for j = 1,2

[z.sub.j] = 1/[g.sub.j] ln (x - [[theta].sub.j]/B/[g.sub.j]) (13)

with [[theta].sub.j] = A - (B/[g.sub.j]) Xj = (B/[absolute value of ([g.sub.j])]). We use the notation X ~ GTMIX(A, B, [g.sub.1], [g.sub.2], [[omega].sub.1]). Vitiello and Poon [19] did not provide the piecewise nature of the mixture density function above in (12). In this case the cdf of X is given by

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (14)

If we assume that U ~ N(0,1), (12) can be written as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII],(16)

where [phi](z) is the standard normal pdf. Note that the expression above matches the pdf of a mixture of three-parameter lognormal distributions. Letting [[theta].sub.1] = [[theta].sub.2] = 0, the above pdf reduces to that of a mixture of two-parameter lognormal distributions.

Given that every normal pdf is a version of the standard normal pdf then if U ~ N([mu], [[sigma].sup.2]) we have

[f.sub.U] (u, [mu], [sigma]) = 1/[sigma] [phi] (u - [mu]/[sigma]), with [mu] [member of] R, [sigma] > 0, (17)

and (12) can be written as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (18)

If the parameters [g.sub.j] are scaled by [sigma], that is, [g.sup.*.sub.j] = [sigma][g.sub.j], then

[z.sub.j] = ln(x - [[theta].sup.*.sub.j]) - [[lambda].sup.*.sub.j]/[g.sup.*.sub.j] (19)

with [[theta].sup.*.sub.j] = A - (B[sigma]/[g.sup.*.sub.j]), [[lambda].sup.*.sub.j] = ln(B[sigma]/[absolute value of ([g.sup.*.sub.j])]) + ([mu]/[sigma])[g.sup.*.sub.j]. Note that the expression above matches the pdf of a mixture of three-parameter lognormal distributions, which is a generalization of the pdf given in (16), and we use the notation X ~ LSMIX([[lambda].sup.*.sub.1], [[lambda].sup.*.sub.2], [g.sup.*.sub.1], [g.sup.*.sub.2], [[theta].sup.*.sub.1], [[theta].sup.*.sub.2], [[omega].sub.1]). Similarly, we can obtain pdf of a mixture of distributions for the random variables listed in Table 1.

4. Estimation of the Mixtures of Two Tukey's g Distributions

In this section, we explain the estimation of the mixture of two Tukey's g distributions. The expected value of X is given by

E [X] = [[mu]'.sub.1] = [2.summation over (j=1)][[omega].sub.j] ([[theta].sub.j] + B/[g.sub.j] Mu ([g.sub.j])]. (20)

The nth raw moment of the random variable X is given by

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (21)

where [g.sub.1][g.sub.2] [not equal to] 0, [??] = (n - k)[g.sub.j] and [M.sub.U](t) is the moment generating function of the random variable U. The central moments [[mu].sub.n] of the random variable X are given by

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (22)

The first five central moments are as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (23)

where for j = 1,2

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (24)

Because [[theta].sub.1] < [[theta].sub.2], upon equating population moments to the corresponding sample moments, it follows from (23) that

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (25)

Left-hand side of system (23) is multiplied by [[omega].sub.1] + [[omega].sub.2] = 1; the equations take the following form:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (26)

where [m.sub.i] (i = 1,2, ...) denote the ith central moment of the sample. Equations (26) accordingly constitute a system of five equations to be solved simultaneously for the estimates of the five parameters A, B, [g.sub.1], [g.sub.2], and [[omega].sub.1].

Note that, from the first equation of system of (26), it follows that

[[omega].sub.1] = [[eta].sub.2]/[[eta].sub.2] - [[eta].sub.1] (27)

We eliminate [[omega].sub.1] between the first and the subsequent equations of (26) in turn and thereby reduce the system to the following four equations in four unknowns A, B, [g.sub.1], and [g.sub.2]:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (28)

These systems of equations are solved computationally by using scientific software package and we do not need to verify the unique solution of the system as the parameter estimates. We skip further details and numerical illustration owing to space constraint.

5. Illustration

In this section we discuss some examples and applications of the results derived in Section 3 with two examples. In the first example, we discuss the pricing of a call option using a mixture of two Tukey's g-generalized distributions as an example to illustrate the results of Section 3. In the second example, we examine the empirical real data of Heating-Degree-Day to demonstrate usefulness of our approach of mixture of LS distributions.

Jimenez et al. [24] derived the option price of an European option assuming that the terminal price distribution follows a g-generalized distribution. Instead if we use a mixture of two Tukey's classes of g-generalized distributions, then the price of the call option denoted by C(t, [tau]; K) with a strike price K and maturity date T = t + [tau] can be expressed as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], (29)

where K > [[theta].sub.2] and

[[delta].sub.j] = [[lambda].sub.j] - ln(K - [[theta].sub.j])/[g.sub.j] for j = 1,2. (30)

When U ~ N(0,1), (29) reduces to

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (31)

where [PHI]* denotes the cdf of a standard univariate normal variable. If we assume that [[theta].sub.1] = [[theta].sub.2] = 0, then (31) reduces to

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (32)

Note that when [g.sub.j] = [[sigma].sub.j] [square root of ([tau])], these expressions coincide with the option pricing formula given in Bahra [25]. The authors also established closed form formula for the calculation of the sensitives measures of option pricing (Greek parameters of the option). Here we wish to observe that our mixture model uses less unknown parameters for calculating the option pricing, whereas Vitiello and Poon [19] used nine unknown parameters to obtain the same for the mixture of two g-distributions. It has been known that when we increase the number of parameters, we lose degrees of freedom and it is no longer acceptable for the best fit of data. This gives an advantage of our approach for the mixture of two g-generalized distributions.

We now present, as an example, the use of Heating-Degree-Days (HDD) in relation to winter temperature risk as a substitute for gas demand. HDD based contracts are listed on the Chicago Mercantile Exchange (CME). We consider an example that consists of monthly aggregate Heating-Degree-Day (HDD) data values at the Chicago O'Hare International Airport from December 1979 to December 2000 given in Wang [26] and explored also by Vitiello and Poon [19]. We describe first a LS distribution with three parameters based method to infer the implied risk-neutral probability density (RND). In Table 3, we present the estimated values of the three parameters of lognormal and Log-Logistic distributions; our interest is to compare with Vitiello and Poon [19] risk-neutral densities with our proposed mixture model.

The smaller value of the Kolmogorov-Smirnov (KS) test confirms that the data obeys the LS distributions with three parameters. We wish to observe that Anderson-Darling (AD) test is more sensitive to the tails of the LS distributions in comparison with KS test. In this case, we choose the Log-Logistic distribution as the best fit for the HDD data.

The implicit risk-neutral densities (RND) of LS distributions are shown in Figure 1 and compared with Figure 6 of Vitiello and Poon [19]. We have obtained a similar plot by our method with less unknown parameters than method given by Vitiello and Poon [19]. Furthermore, their KS test value of 13.6326% which is higher than the KS test values of Table 3 favors the best fit for the frequency of the LS distributions. Therefore, finite mixtures are attractive from the application viewpoint because of its flexibility and permit us to model various kinds of shaped distributions. In Table 4, we give the estimate values of the parameters of the mixture LS distributions. These parameters are estimated using (28). The estimated two g-densities and the implied risk-neutral densities (RND) are shown in Figure 2.

We observe that the bimodal LS mixture distribution has same fitting

performance of the empirical distribution function (EDF) and lognormal mixture distribution gives best goodness of fit using the KS test.

6. Conclusions

This paper presents a mixture of Tukey's g-generalized distributions and its properties. The methodology of estimating the unknown parameters by the method of moments is also presented. The proposed model has the advantage that it provides flexibility, when skewness, kurtosis, or other moments of the underlying distribution do not follow a normal distribution. Some special cases of well known distributions are obtained from the proposed model.

http://dx.doi.org/10.1155/2016/3509139

Competing Interests

The authors declare that they have no competing interests.

References

[1] B. Mandelbrot, "The variation of certain speculative prices," The Journal of Business, vol. 36, no. 4, pp. 394-419,1963.

[2] E. F. Fama, "The behavior of stock-market prices," The Journal of Business, vol. 38, no. 1, pp. 34-105,1965.

[3] S. J. Kon, "Models of stock returns--a comparison," The Journal of Finance, vol. 39, no. 1, pp. 147-165,1984.

[4] Y. K. Tse, "Price and volume in the tokyo stock exchange: an exploratory study," Working Paper 1573, University of Illinois at Urbana-Champaign, 1989.

[5] B. D. Fielitz and J. P. Rozelle, "Stable distributions and the mixtures of distributions hypotheses for common stock returns," Journal of the American Statistical Association, vol. 78, no. 381, pp. 28-36,1983.

[6] J. A. Jimenez, V. Arunachalam, and G. M. Serna, "Option pricing based on a log-skew-normal mixture," International Journal of Theoretical and Applied Finance, vol. 18, no. 8, Article ID 1550051, 22 pages, 2015.

[7] J. W. Tukey, "Modern techniques in data analysis," in Proceedings of the NSF Sponsored Regional Research Conference, Southeastern Massachusetts University, North Dartmouth, Mass, USA, 1977.

[8] J. Martinez and B. Iglewicz, "Some properties of the tukey g and h familyof distributions," Communications in Statistics--Theory and Methods, vol. 13, no. 3, pp. 353-369,1984.

[9] S. G. Badrinath and S. Chatterjee, "On measuring skewness and elongation in common stock return distributions: the case of the market index," The Journal of Business, vol. 61, no. 4, pp. 451-472, 1988.

[10] S. G. Badrinath and S. Chatterjee, "A data-analytic look at skewness and elongation in common-stock-return distributions," Journal of Business & Economic Statistics, vol. 9, no. 2, pp. 223-233, 1991.

[11] T. C. Mills, "Modelling skewness and kurtosis in the london stock exchange ft--se index return distributions," The Statistician, vol. 44, no. 3, pp. 323-332,1995.

[12] K. K. Dutta and D. F. Babbel, On Measuring Skewness and Kurtosis in Short Rate Distributions: The Case of the US Dollar London Inter Bank Offer Rates, The Warton School, University of Pennsylvania, Philadelphia, Pa, USA, 2004.

[13] K. K. Dutta and D. F. Babbel, "Extracting probabilistic information from the prices of interest rate options: tests of distributional assumptions," Journal of Business, vol. 78, no. 3, pp. 841870, 2005.

[14] X. Tang and X. Wu, "A new method for the decomposition of portfolio VaR," Journal of Systems Science and Information, vol. 4, no. 4, pp. 721-727, 2006.

[15] K. K. Dutta and J. Perry, "A tale of tails: an empirical analysis of loss distribution models for estimating operational risk capital," Working Paper 06-13, Federal Reserve Bank of Boston, 2007

[16] J. A. Jimenez and V. Arunachalam, "Evaluating operational risk by an inhomogeneous counting process based on Panjer recursion," The Journal of Operational Risk, vol. 11, no. 1, pp. 1-21, 2016.

[17] J. A. Jimenez and V. Arunachalam, "Using Tukey's g and h family of distributions to calculate value at risk and conditional value at risk," Journal of Risk, vol. 13, no. 4, pp. 95-116, 2011.

[18] J. A. Jimenez, V. Arunachalam, and G. M. Serna, "A generalization of Tukey's g-h family of distributions," Journal of Statistical Theory and Applications, vol. 14, no. 1, pp. 28-44, 2015.

[19] L. Vitiello and S.-H. Poon, "General equilibrium and risk neutral framework for option pricing with a mixture of distributions," The Journal of Derivatives, vol. 15, no. 4, pp. 48-60,2008.

[20] D. C. Hoaglin and S. C. Peters, Software for Exploring Distribution Shape, Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, 1979.

[21] N. L. Johnson, S. Kotz, and N. Balakrishnan, Continuous Univariate Distributions, vol. 1, John Wiley & Sons, New York, NY, USA, 1994.

[22] A. Stuart and J. K. Ord, Kendall's Advanced Theory of Statistics: Distribution Theory, vol. 1, Edward Arnold, London, UK; Halsted Press, New York, NY, USA, 6th edition, 1994.

[23] D. M. Titterington, A. F. Smith, and U. E. Makov, Statistical Analysis of Finite Mixture Distributions, vol. 7, John Wiley & Sons, New York, NY, USA, 1985.

[24] J. A. Jimeenez, V. Arunachalam, and G. M. Serna, "Option pricing based on the generalised Tukey distribution," International Journal of Financial Markets and Derivatives, vol. 3, no. 3, pp. 191-221, 2014.

[25] B. Bahra, "Implied risk-neutral probability density functions from option prices: a central bank perspective," in Forecasting Volatility in the Financial Markets, J. Knight and S. Satchell, Eds., pp. 201-226, Elsevier, 3rd edition, 1997

[26] S. S. Wang, "A universal framework for pricing financial and insurance risks," ASTIN Bulletin, vol. 32, no. 2, pp. 213-234, 2002.

Jose Alfredo Jimenez and Viswanathan Arunachalam

Department of Statistics, Universidad Nacional de Colombia, Carrera 45 No. 26-85, Bogota, Colombia

Correspondence should be addressed to Viswanathan Arunachalam; varunachalam@unal.edu.co

Received 8 April 2016; Revised 1 July 2016; Accepted 14 July 2016

Academic Editor: Chin-Shang Li

Caption: Figure 1: Empirical and LS([LAMBDA]) densities estimated from HDD.

Caption: Figure 2: Empirical and two-g densities estimated from HDD.
TABLE 1: Parameters of the pdf of the random variable Z = ln(X).

Distribution               Parameters
of the r.v. U
                  [mu], a            [sigma], b

Laplace              0         [square root of (2)]/2

Logistic             0        [square root of (3)]/[pi]

Normal               0                    1

HyperSec             0                 2/[pi]

HyperCsc             0        [square root of (2)]/[pi]

Distribution             Parameters
of the r.v. U
                     g [not equal to] 0

Laplace          [absolute value of (g)] <
                  [square root of (2)]/[pi]

Logistic         [absolute value of (g)] <
                 [pi]/[square root of (2)]n

Normal                 g [member of] R

HyperSec         [absolute value of (g)] <
                 [pi]/[square root of (2)]n

HyperCsc          [absolute value of (g)] <
                 [pi]/[square root of (2)]n

Distribution    Distribution
of the r.v. U   of the r.v. Z

Laplace          Log-Laplace

Logistic        Log-Logistic

Normal            Lognormal

HyperSec         LoghyperSec

HyperCsc         LoghyperCsc

Distribution                        Parameters
of the r.v. U
                        [mu], a                      a, b

Laplace                  ln(B/              [square root of (2)]/2
                [absolute value of (g)])    [absolute value of (g)]

Logistic                 ln(B/             [square root of (3)]/[pi]
                [absolute value of (g)])    [absolute value of (g)]

Normal                   ln(B/              [absolute value of (g)]
                [absolute value of (g)])

HyperSec                 ln(B/                      2/[pi]
                [absolute value of (g)])    [absolute value of (g)]

HyperCsc                 ln(B/             [square root of (2)]/[pi]
                [absolute value of (g)])    [absolute value of (g)]

TABLE 2: Parameters of the pdf and moment generating functions of
the random variable U.

Distribution             Parameters
of the r.v. U
                [mu], a          [sigma], b

Laplace            0       [square root of (2)]/2

Logistic           0      [square root of (3)]/[pi]

Normal             0                  1

HyperSec           0               2/[pi]

HyperCsc           0      [square root of (2)]/[pi]

Distribution             Parameters
of the r.v. U
                     g [not equal to] 0

Laplace          [absolute value of (g)] <
                  [square root of (2)]/[pi]

Logistic         [absolute value of (g)] <
                 [pi]/[square root of (2)]n

Normal                 g [member of] R

HyperSec         [absolute value of (g)] <
                 [pi]/[square root of (2)]n

HyperCsc          [absolute value of (g)] <
                 [pi]/[square root of (2)]n

Distribution
of the r.v. U         [M.sub.U](g)

Laplace              2/2 - [g.sup.2]

Logistic        [square root of (3)]g csc
                 ([square root of (3)]g)

Normal             exp {1/2[g.sup.2]}

HyperSec                 sec(g)

HyperCsc               [sec.sup.2]
                (g/[square root of (2)])

TABLE 3: Estimates for adjusting the LS(A).

Distribution of               Parameters            Test of adjusted
three parameters
                   [lambda]      g       [theta]    AD (%)   KS (%)

Lognormal           5.7407     0.5265    798.2540   37.53    12.3041
Log-Logistic        3.0225    283.8185   824.1814   32.68    11.1367

TABLE 4: Estimates for adjusting the mixture of LS(A).

Mixture of                     Parameters
distributions
                [[lambda].sub.1];        [g.sub.1];
                [[lambda].sub.2]         [g.sub.2]

Lognormal            6.5749               -0.1268

                     11.0933        5.7574 x [10.sup.-4]
                     10.1338             -538.2433

Log-Logistic
                    303.7429             6818.2894

Mixture of         Parameters
distributions
                [[theta].sub.1];
                [[theta].sub.2]

Lognormal          1797.9995

                  -64213.6284
                   1620.7711

Log-Logistic
                   -5309.4256

Mixture of
distributions             KS (%) test
                [omega]

Lognormal
                0.8188      8.0866

Log-Logistic    0.8182      8.5928
COPYRIGHT 2016 Hindawi Limited
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Research Article
Author:Jimenez, Jose Alfredo; Arunachalam, Viswanathan
Publication:Journal of Probability and Statistics
Article Type:Report
Date:Jan 1, 2016
Words:4157
Previous Article:Confidence region approach for assessing bioequivalence and biosimilarity accounting for heterogeneity of variability.
Next Article:Odds ratios estimation of rare event in binomial distribution.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters