sibsutisВестник СибГУТИThe Herald of the Siberian State University of Telecommunications and Information Science1998-6920СибГУТИsibsutis-262Research ArticleСтатьиМощность критерия однородности как функция полезности в задачах принятия решения в условиях риска и неопределенностиHomogeneity test power as utility function in the theory of decision making under risk and uncertaintyФилоненкоП. А.PhilonenkoP. ..petr-filonenko@mail.ruПостоваловС. Н.PostovalovS. ..postovalov@ngs.ruНГТУRussian Federation20172410202203320Copyright © Филоненко П.А., Постовалов С.Н., 20222022Филоненко П.А., Постовалов С.Н.Philonenko P..., Postovalov S...Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://vestnik.sibsutis.ru/jour/article/view/262

В задачах проверки статистических гипотез существует большое количество статистических методов. Некоторые статистические критерии являются предпочтительными при определенных альтернативных гипотезах. Необходим некоторый надежный способ выбора предпочтительного критерия. Мы формируем типы альтернативных гипотез (с различным количеством точек пересечений функций надежности), для каждого типа строим несколько альтернативных гипотез (с разными законами распределения моментов отказа), а затем в соответствие с теорией принятия решений в условиях риска и неопределенности (критерий Вальда) мы получаем, какой критерий предпочтительнее при определенном типе альтернативной гипотезы, используя результаты компьютерного моделирования мощности критериев.

There are a lot of statistical tests for hypothesis testing and some statistical tests are preferable than others in a certain alternative hypothesis. It needs some reliable method for selecting the powerful statistical test. To solve the problem, we create types of alternative hypotheses (with different number of intersection points of reliability function). For every type we create some alternative hypotheses (with various distribution of survival function) with similar behavior of survival functions (the behavior of test power is similar too) to simulate the power of statistical tests and then we apply the Wald test to conclude what test is the most preferable in a certain type of alternative hypotheses.

проверка статистических гипотезгипотеза однородности распределенийданные типа времени жизникритерий Вальдаметод Монте-Карлоtwo-sample problemlifetime dataWald testMonte-Carlo methodReferencesPhilonenko P., Postovalov S. A power comparison of homogeneity tests for randomly censored data / P. Philonenko, S. Postovalov // Applied methods of statistical analysis. Applications in survival analysis, reliability and quality control AMSA2013, Novosibirsk, 25-27 Sept. 2013: proc. of the intern. workshop. Novosibirsk: NSTU publ., 2013. P. 227-237.Philonenko P., Postovalov S. A power comparison of homogeneity tests for randomly censored data / P. Philonenko, S. Postovalov // Applied methods of statistical analysis. Applications in survival analysis, reliability and quality control AMSA2013, Novosibirsk, 25-27 Sept. 2013: proc. of the intern. workshop. Novosibirsk: NSTU publ., 2013. P. 227-237.Philonenko P., Postovalov S. A Comparison of Homogeneity Tests for Different Alternative Hypotheses // Statistical Models and Methods for Reliability and Survival Analysis: monograph. London: Wiley-ISTE, 2013. Chap. 12. P. 177-194.Philonenko P., Postovalov S. A Comparison of Homogeneity Tests for Different Alternative Hypotheses // Statistical Models and Methods for Reliability and Survival Analysis: monograph. London: Wiley-ISTE, 2013. Chap. 12. P. 177-194.Wald A. Statistical decision functions which minimize the maximum risk. The Annals of Mathematics. 1945. V. 46, № 2. P. 265-280.Wald A. Statistical decision functions which minimize the maximum risk. The Annals of Mathematics. 1945. V. 46, № 2. P. 265-280.Johnson E. J., Payne J. W. Effort and accuracy in choice. Management Science. 1985. V. 31, № 4. P. 395-414.Johnson E. J., Payne J. W. Effort and accuracy in choice. Management Science. 1985. V. 31, № 4. P. 395-414.Wilcoxon F. Individual comparisons by ranking methods. Biometrics Bulletin. 1945. V. 1, № 6. P. 80-83.Wilcoxon F. Individual comparisons by ranking methods. Biometrics Bulletin. 1945. V. 1, № 6. P. 80-83.Lee E. T., Wang J. W. Statistical Methods for Survival Data Analysis. N. Y: Wiley, 2003.Lee E. T., Wang J. W. Statistical Methods for Survival Data Analysis. N. Y: Wiley, 2003.Gehan E. A. A generalized Wilcoxon test for comparing arbitrarily singly-censored samples // Biometrika. 1965. V. 52, № 1/2. P. 203-223.Gehan E. A. A generalized Wilcoxon test for comparing arbitrarily singly-censored samples // Biometrika. 1965. V. 52, № 1/2. P. 203-223.Peto R., Peto J. Asymptotically efficient rank invariant test procedures // J. Royal Statist. Soc. Ser. A (General). 1972. V. 135, № 2. P. 185-207.Peto R., Peto J. Asymptotically efficient rank invariant test procedures // J. Royal Statist. Soc. Ser. A (General). 1972. V. 135, № 2. P. 185-207.Kaplan E. L. and Meier P. Nonparametric estimator from incomplete observation // J. Amer. Statist. Assoc. 1958. V. 53. P. 457-481.Kaplan E. L. and Meier P. Nonparametric estimator from incomplete observation // J. Amer. Statist. Assoc. 1958. V. 53. P. 457-481.Savage I. R. Contributions to the Theory of Rank Order Statistics: The Two Sample Case // Annals of Mathematical Statistics. 1956. V. 27. P. 590-615.Savage I. R. Contributions to the Theory of Rank Order Statistics: The Two Sample Case // Annals of Mathematical Statistics. 1956. V. 27. P. 590-615.Mantel N. Evaluation of Survival Data and Two New Rank Order Statistics Arising in Its Consideration // Cancer Chemotherapy Reports. 1966. V. 50. P. 163-170.Mantel N. Evaluation of Survival Data and Two New Rank Order Statistics Arising in Its Consideration // Cancer Chemotherapy Reports. 1966. V. 50. P. 163-170.Pepe M. S., Fleming T. R. Weighted Kaplan-Meier statistics: A class of distance tests for censored survival data // Biometrics. 1989. V. 45. P. 497-507.Pepe M. S., Fleming T. R. Weighted Kaplan-Meier statistics: A class of distance tests for censored survival data // Biometrics. 1989. V. 45. P. 497-507.Ruvie Lou Maria C. Martinez, Joshua D. Naranjo. A pretest for choosing between logrank and wilcoxon tests in the two-sample problem // International Journal of Statistics. 2010. V. LXVIII, № 2, P. 111-125.Ruvie Lou Maria C. Martinez, Joshua D. Naranjo. A pretest for choosing between logrank and wilcoxon tests in the two-sample problem // International Journal of Statistics. 2010. V. LXVIII, № 2, P. 111-125.Philonenko P., Postovalov S. A new two-sample test for choosing between log-rank and Wilcoxon tests with right-censored data // Journal of Statistical Computation and Simulation. 2015. V. 85, № 14. P. 2761-2770.Philonenko P., Postovalov S. A new two-sample test for choosing between log-rank and Wilcoxon tests with right-censored data // Journal of Statistical Computation and Simulation. 2015. V. 85, № 14. P. 2761-2770.Philonenko P., Postovalov S., Kovalevskii A. The limit test statistic distribution of the maximum value test for right-censored data // Journal of Statistical Computation and Simulation. 2016. V. 86, № 17. P. 3482-3494.Philonenko P., Postovalov S., Kovalevskii A. The limit test statistic distribution of the maximum value test for right-censored data // Journal of Statistical Computation and Simulation. 2016. V. 86, № 17. P. 3482-3494.Welch B. L. The generalization of Student's problem when several different population variances are involved // Biometrika. 1947. V. 34. P. 29-35.Welch B. L. The generalization of Student's problem when several different population variances are involved // Biometrika. 1947. V. 34. P. 29-35.Student. The probable error of a mean. // Biometrika. 1908. V. 6, № 1. P. 1-25.Student. The probable error of a mean. // Biometrika. 1908. V. 6, № 1. P. 1-25.Bagdonavicus V. B., Levuliene R. J., Nikulin M. S. Zdorova-Cheminadeo “Tests for equality of survival distributions against non-location alternatives” // Lifetime Data Analysis. 2004. V. 10, № 4. P. 445-460.Bagdonavicus V. B., Levuliene R. J., Nikulin M. S. Zdorova-Cheminadeo “Tests for equality of survival distributions against non-location alternatives” // Lifetime Data Analysis. 2004. V. 10, № 4. P. 445-460.Bagdonavicus V. B., Nikulin M. On goodness-of-fit tests for homogeneity and proportional hazards // Applied Stochastic Models in Business and Industry. 2006. V. 22, № 1. P. 607-619.Bagdonavicus V. B., Nikulin M. On goodness-of-fit tests for homogeneity and proportional hazards // Applied Stochastic Models in Business and Industry. 2006. V. 22, № 1. P. 607-619.Bagdonavicus V. B., Kruopis J., Nikulin M. S. Nonparametric tests for censored data. John Wiley & Sons, Inc., New York, 2010, 233 p.Bagdonavicus V. B., Kruopis J., Nikulin M. S. Nonparametric tests for censored data. John Wiley & Sons, Inc., New York, 2010, 233 p.Bagdonavicus V. B., Kruopis J., Nikulin M. S. Censored and Truncated Data, in Non-parametric Tests for Censored Data. John Wiley & Sons, Inc, Hoboken, NJ, USA. 2013.Bagdonavicus V. B., Kruopis J., Nikulin M. S. Censored and Truncated Data, in Non-parametric Tests for Censored Data. John Wiley & Sons, Inc, Hoboken, NJ, USA. 2013.Titchmarsh E.C. The theory of functions. Oxford University Press, 1976.Titchmarsh E.C. The theory of functions. Oxford University Press, 1976.Kolmogorov A. Some Works of Recent Years in the Field of Limit Theorems in the Theory of Probability // Vestnik Moskov, Univ. Ser. Fiz.-Mat. Estest. Nauk. 1953. V. 8. P. 29-38.Kolmogorov A. Some Works of Recent Years in the Field of Limit Theorems in the Theory of Probability // Vestnik Moskov, Univ. Ser. Fiz.-Mat. Estest. Nauk. 1953. V. 8. P. 29-38.

The authors declare that there are no conflicts of interest present.