Abstract
As it is the key chapter on software development, it begins with a characterization of bivariate normal distribution by Ludeman (Random processes, filtering, estimation and detection, Wiley India, New Delhi, 2010). That is followed by a brief presentation of the various properties of bivariate normal distribution and its applications by Essenwagner (Applied statistics in atmospheric science, part A. Frequencies and curve fitting, Elsevier, New York, 1976). Thereafter, Owen’s (1956) computational scheme for numerical integration of the bivariate normal integral is presented stepwise. The envisaged algorithm is to make iterative use of this scheme. It is very well realized that unlike univariate normal, which has a unique quantile value for a given probability level, its bivariate extension would have multiple (or infinite) quantile pairs for the same probability level. Apart from this, there arose other problems in their generation, which were sorted out and strategies to meet such problems were listed and used in perfecting the same algorithm. This was followed by finding the role of equi-quantile value BIGH for each probability level and correlation value as the initial point of such iterative scheme for the entire computational horizon of four hundred grids. In fact, the approach adopted is entirely innovative and of great economic and other consequences. Such action resulted in expansion of decision alternatives for the given or estimated correlation value without any change in probability (risk) level. Methods of forming simultaneous (joint) confidence intervals have emanated by using such biquantile pairs, having a definite edge over the existing Bonferroni’s joint confidence interval. Multiplicity of biquantile pairs offers a scope even for multiple joint confidence intervals for the same confidence probability. That could yield larger number of decision alternatives and hence the scope of choice amongst them by the decision maker on some criterion function. Those who are interested only in application of quantile pairs may skip Sect. 4.1, but they also should read its concluding part.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Also discussed in Sect. 3.1 (in Chap. 3).
- 2.
For further theoretical background, Ludeman’s text, or that of Papoulis (1965), should be consulted.
- 3.
The impact of these shall be seen in examples in Chap. 7.
- 4.
Further, their illustrations through varied examples are given of Chap. 7.
- 5.
- 6.
His exposition of bivariate elliptical distribution, an outcome of non-zero correlation between x 1 and x 2, where x 1 (zonal wind speed data) and x 2 (meridional wind speed data), shall be discussed in Sect. 7.5 of Chap. 7.
- 7.
- 8.
It was not considered necessary to compare the results so obtained with those of others because results obtained met all the software reliability test criteria as presented in Chap. 5.
- 9.
As discussed in Chap. 3.
- 10.
Some of which are discussed in Chap. 7.
- 11.
As mentioned in Sect. 3.5 in Chap. 3.
- 12.
Some of which have been illustrated through numerous examples in Chap. 7.
- 13.
- 14.
Thus the problem of the size of the two-way table having been determined, its 21 columns for the range of the correlation values between +0.95 and −0.95 are required to be spread in three sheets of Table 8.1 of Chap. 8 (seven correlation values in descending order of magnitude in a sheet), whereas its rows are required to accommodate all the 19 probability levels between 0.99 and 0.50, common for all the three sheets of the said Table 8.1 with intervals as decided above. Thus, there could be 21 columns for differing correlation values and 19 rows for changing probability levels having 21 × 19 = 399 grids for which iso-probable quantile (biquantile) pair tables are required to be generated.
- 15.
Such a table of 399 grids, providing suitable data structure, is placed in Table 8.1 of Chap. 8. Its three-dimensional graph obtained by the use of MATLAB 7 with some changes in data to meet its programming requirements will be presented on a separate sheet following this page to illustrate its features.
- 16.
It has been placed at the said Table 8.1 of Chap. 8 along with the corresponding SMHL values just below SMH in the same cell.
- 17.
It is discussed in Sect. 3.5 in Chap. 3.
- 18.
Refer to the illustration given in Sect. 7.4 in Chap. 7.
- 19.
However, a negative correlation has been used in more than one example given in Sects. 7.1, 7.3 and 7.9 in Chap. 7, from which its uses and implications can easily be understood. The earliest version of the software BIVNOR, the software for generating tables of bivariate normal quantile pairs was presented by Das (1993) at the 46th Annual Conference of I.S.A.S. held at Bhubaneshwar in February 1993.
- 20.
Numerical examples taken from Johnson and Wichern (op. cit.) comparing all the above methods with joint confidence intervals based on biquantile pairs have been presented in Sect. 7.2 of Chap. 7).
- 21.
Please refer to http://www.biostat.wustl.edu/archives/html/s-news/2003-2msg00004.html. This monographic text and especially the tables generated and placed in Chap. 8, which are Tables 8.1 and 8.2 are the answers to David’s questions. It is also advised to read sections of Chap. 7 and descriptions on Tables in Chap. 5.
- 22.
Refer to “quantile-bivariate-normal distribution” posted on Tuesday 20 December 2005, 02:37:30 GMT.
- 23.
However, the abstract is released to http://atlas-conferences do not throw any light on the role of correlation values on the estimates of quantiles. Hopefully, they could have done so. Though the said work appears to be nearer to the present one, it throws no light on the actual generation of quantile pairs for a given probability level and known or estimated correlation values as has been obtained in Chap. 8 of the book.
- 24.
Also presented in Table 8.1 of Chap. 8.
- 25.
References
Armstrong, R.D., Balintfy, J.L.: A chance-constrained multiple choice programming algorithm. Oper. Res. 23, 494–510 (1975)
Balintfy, J.L.: Nonlinear programming for models with joint chance-constraints. In: Integer and Nonlinear Programming, pp. 337–352. Elsevier, New York (1970)
Beasley, J.D., Springer, S.G.: Algorithm AS 111: the percentage points of the normal distribution. Appl. Statist. 26, 118–121 (1977)
Bereanu, B.: Stochastic programming. Rev. Math. Pures et Appl. 4 (1963a)
Bereanu, B.: Stochastic transportation problem I: random consumption. Com. Acad. R.P.R 4, 325–331 (1963b)
Bereanu, B.: On stochastic transportation problem II: random consumption. Com. Acad. R.P.R, 332–337 (1963c)
Bonferroni’s C.E.: Publication not reported (1892–1960)
Borth, D.M.: A modification of Owen’s method for computing the bivariate normal integral, Appl. Stat. 22, 82–85 (1973)
Bradley, R.A.: Multivariate Analysis I: overview. In: Kruskal, W.H., Tanur, J.M. (eds.) International Encylopedia of Statistics. Free Press, Newyork (1978)
Cadwell, J.H.: The bivariate normal integral. Biometrika 38, 475–479 (1951)
Cramer, H.: Mathematical Methods of Statistics. Princeton University Press, Princeton (1951)
Das, N.C.: On software for generating tables of bivariate normal quantile pairs. J. Ind. Soc. Agric. Stat. 45(1), 130 (1993)
David Shin C.: Bivariate normal quantile function. Minbucket in part, second question p. 3 of 3, 2 Nov 2008, and p. 1 of 2, 26 Feb 2008 (2003)
Deak, I.: Computation of multiple normal probabilities. In: Prekopa, A., Kall, P. (eds.). Symposium on Stochastic Programming. Springer’s Lecture Notes in Economics and Mathematical Systems, pp. 107–20 (1979)
Deak, I.: Multidimensional integration and stochastic. Wets, R. (ed.) Proceedings of the Workshop in Numerical Methods for Stochastic Optimization. IIASSA, Luxemburg (1983)
Deak, I.: Subroutine for computing normal probabilities of sets: computer Experiences. Ann. Oper. Res. 100, 103–122 (2000)
Digvi, D.R.: Calculation of univariate and bivariate normal probability function. Ann. Stat. 7(4), 903–910 (1975)
Essenwagner, O.: Applied Statistics in Atmospheric Science, Part A. Frequencies and Curve Fitting. Elsevier, New York (1976)
Galton, F.: Natural Inheritance. MacMillan, New York (1886)
Galton, F.: Memories of My Life. Methuen, London (1908)
Genz, A., Brez, F.: Computations of multivariate normal and t probabilities. Lecture Notes in Statistics, vol. 195. Springer-verlag, Heidelberg (2009)
Genz, A., Brez, F.: Internet google search. http://www.sci.wsu.edu/math/faculty/genz/homepage (2012) Feb 2012
Genz, A., Bretz, F., Miwa, T., Mi, X., Leisch, F., Scheipl, F., Hothorn, T.: http://www.sci.wsu.edu/math/faculty/genz/homepage (2010)
Genz, A., Bretz, F., Miwa, T., Mi, X., Leisch, F., Scheipl, F., Hothorn, T.: With Maintainer, Torsten Hothorn. Torsten.Hothorn@R-project.org (2012)
Gupta, Shanti S.: Probability integrals of multivariate normal and multi-variate—t. Ann. Math. Stat. 34, 792–828 (1963a)
Gupta, Shanti S.: Bibliography on multivariate normal integrals and related topics. Ann. Math. Stat. 34, 829–837 (1963b)
Hajivassiliou, V., McFadden, D., Ruud, P.: Simulation of multivariate normal rectangle probabilities and their derivatives. J. Econo. 72, 85–114 (1996)
Hill, I.D.: The normal integral. In: Griffiths, P., Hill, I.D. (eds.) Application Statistics. Algorithms. Ellis Herwood Limited and the Royal Statistical Society, London (1985)
Hull, J.C.: Risk Management and Financial Institutions. Pearson Education, New Delhi (2007)
Hull, J.C.: Options, Futures and other Derivatives. Pearson Education, New Delhi (2009)
Hymans, S.H.: Simultaneous confidence intervals in economic forecasting. Econometrica 36, 18–30 (1968)
Jagannathan, R., Rao, M.R.: A class of nonlinear chance-constrained programming models with joint constraint. Oper. Res. 21(1), 360–364 (1973)
Jagannathan, R.: Chance-constrained programming models with joint chance constraints. Oper. Res. 22(2), 358–372 (1974)
Johnson, R.A., Wichern, D.W.: Applied Multivariate Statistical Analysis. PHI Learning, New Delhi (2003)
Keane, M.: A computationally practical simulation estimator for panel data. Econometrica 62, 96–116 (1994)
Kerridge, D.F., Cook, G.W.: Yet another series for normal integral. Biometrika 63, 401–403 (1976)
Kumar, S., Tripathy, M.R.: Estimating quantile of normal Population. http://atlas-conferences.com/c/a/t/a/45.html (2007)
Lo, S.M.S., Wilke, R.A.: A copula model for dependent competing risks. J. Roy. Stat. Soc. Series C. (Appl. Stat.) 59(2), 359–376 (2010)
Ludeman, L.C.: Random Processes, Filtering, Estimation and Detection. Wiley India, New Delhi (2010)
Majumdar, K. L., Bhattacharjee G.P.: Algorithm AS64/AS104: inverse of incomplete beta function ratio. In Griffiths, P., Hill, I.D. (eds.) Applied Statistics Algorithm. Ellis Horwood Ltd. For the Royal Statistical Society, London (1973, 1985/86)
Marshall, A.W., Olkin, I.: Inequalities: Theory of Majorization and its Applications. Academic Press, New York (1979)
Millar Jr, R.G.: Simultaneous Statistical Inference. McGraw Hill, New York (1966)
Milton, J.: Paradise Lost. New Delhi (2009, 1667)
Milton, R.C., Hotchkiss, R.: Computer evaluation of the normal and inverse normal distribution function. Technometrics 11, 817–822 (1969)
Milton, R.C.: Computer evaluation of the multivariate normal integral. Technometrics 14(4), 881–889 (1972)
Moran, P.A.P.: Calculation of normal distribution function. Biometrika 67(3), 675–676 (1980)
National Bureau Standards. U.S.A (1959)
Neyman, J.: On the problem of confidence intervals. Ann. Math. Statis. 6, 111 (1935)
Neyman, J.: Fiducial argument and theory of confidence intervals. Biometrika 32, 128 (1941)
Nicholson, C.: The probability integral of two variables. Biometrika 33, 59–72 (1943)
Odeh, R.E., Evans, J.O.: Algorithm AS 70: the percentage points of the normal distribution. Appl. Stat. 23, 96–97: Ann. Math. Statis. 27, 1075–1090 (1974)
Owen, Donald B.: Tables for computing bivariate normal probabilities. Ann. Math. Statist. 27, 1075–1090 (1956)
Papoulis, A.: Probability, Random Variables and Stochastic Processes. McGraw-Hill Kogakusha Ltd. Tokyo, London (1965)
Pearson, K.: Karl Pearson’s Early Statistical Papers. Cambridge University Press, Cambridge (1900:1948)
Pearson, K.: The Tables for Statisticians and Biometricians, vol. 2. University College, London (1914:1930–31)
Prekopa, A.: Logarithmic concave measures with application to stochastic programming. Acta Scientiarum Mathematicarum 32, 301–316 (1971)
Prekopa, A.: Contribution to the theory of stochastic programming. Math. Prog. 4, 202–221 (1973)
Prekopa, A., Szantai, T.: On optimal regulation of a storage level with application to the water level regulation of a lake: a survey of mathematical programming. In: Proceedings of the 9th International Mathematical Program Symposium, pp. 183–210. Akademiai Kiado, Budapest (1976)
Prekopa, A., Kelle, P.: Reliability type inventory models based on stochastic programming. Math. Prog. Study. 9, 43–58. North Holland Publishing Company, Amsterdam (1978)
Rizopoulos, D.: Quantiles for bivariate normal distribution. https://www.R-project.org/posting-guide.html http://www.r-project.org/posting-guide.html (2009)
Rose, C., Smith, M.D.: Bivariate normal, section 6.4A. In: Mathematical Statistics with Mathematica. Springer Text in Statistics. Springer, Berlin (2006)
Rossi, P.E., Allenby, G.M., McCulloch, R.: Bayesian Statistics and Marketing. Wiley, New York (2005)
Royhaas. Answer quoted in the text. S.O.S. Mathematics Cyber Board. quantile-bivariate norm.dist (2005)
Roy, S.N., Bose, R.C.: Simultaneous confidence interval estimation. Ann. Math. Stat. 24, 5133–5136 (1953)
Roy, S.N.: Some further results in simultaneous confidence interval estimation. Ann. Math. Stat. 25, 752–761 (1954)
Roy, S.N.: Some Aspects of Multivariate Analysis. Asia Publishing House, New Delhi; Indian Statistical Institute Calcutta, Kolkata (1958)
Sengupta, J.K., Tintner, G., Millham, C.: On some theorems on stochastic linear programming with applications. Management Sci. 10, 143–159 (1963)
Sengupta, J.K.: Safety first rules under chance constrained linear programming. Oper. Res. 17, 112–132 (1969)
Sengupta, J.K.: Stochastic Programming: Methods and Applications. North Holland Publishing Company, Amsterdam (1971)
Sengupta, J.K.: Decision Models in Stochastic Programming. North Holland Publishing Company, Amsterdam (1982)
Sheppard, W.F.: On the calculation of double integral expressing normal correlation. Trans. Camb. Phil. Soc. 19, 23–69 (1900)
Sheppard, W.F.: The probability integral. In: British Association Mathematical Tables, vol. 7. Cambridge University Press, Cambridge (1939, posthumous)
Sheppard, W.F.: Philos. Trans. Roy. Soc. Lond. (A) 192, 101–167 (1999)
Symonds, G.: Chance-constrained equivalents of some stochastic programming problems. Oper. Res. 16, 1152–1159 (1968)
Vasicek, O.: An equilibrium characterization of the term structure. J. Finan. Econ. 5, 177–188 (1977)
Willink, R.: Bounds on the bivariate normal distribution function. Comm. Stat. Theory Methods 33(10), 2281–2297 (2004)
Wolfram S.: Mathematica, Ver. 3, 3rd ed. Wolfram Media, Melbourne; Cambridge University Press, Cambridge (1996)
Young, J.C., Minder, Ch.E.: Algorithm AS-76: an integral useful in calculating non-central t and bivariate normal probabilities. In: Griffiths, P., Hills, I.D. (eds.) Applied Statistics Algorithm, pp. 145–48. [op. cit.] (1985)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2015 Springer India
About this chapter
Cite this chapter
Das, N.C. (2015). Bivariate Normal Distribution and Heuristic-Algorithm of BIVNOR for Generating Biquantile Pairs. In: Decision Processes by Using Bivariate Normal Quantile Pairs. Springer, New Delhi. https://doi.org/10.1007/978-81-322-2364-1_4
Download citation
DOI: https://doi.org/10.1007/978-81-322-2364-1_4
Published:
Publisher Name: Springer, New Delhi
Print ISBN: 978-81-322-2363-4
Online ISBN: 978-81-322-2364-1
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)