Constancy of distributions: nonparametric monitoring of probability distributions over time
نویسندگان
چکیده
In this paper we study stochastic processes which enable monitoring the possible changes of probability distributions over time. These processes may in particular be used to test the null hypothesis of no change. The monitoring processes are bivariate functions, of time and position at the measurement scale, and are approximated with zero mean Gaussian processes under the constancy hypothesis. One may then form Kolmogorov–Smirnov or other type of tests as functionals of the processes. To study null distributions of the resulting tests, we employ KMT-type inequalities to derive Cramér-type deviation results for (bootstrapped versions of) such tests statistics.
منابع مشابه
A comparison of parametric and non-parametric methods of standardized precipitation index (SPI) in drought monitoring (Case study: Gorganroud basin)
The Standardized Precipitation Index (SPI) is the most common index for drought monitoring. Although the calculation of this index is usually done by using the gamma distribution fitting of precipitation data, studies have shown that for accurate monitoring of drought, the optimal distribution of precipitation in each month should be determined. On the other hand, in non-stationary time series,...
متن کاملIntroducing of Dirichlet process prior in the Nonparametric Bayesian models frame work
Statistical models are utilized to learn about the mechanism that the data are generating from it. Often it is assumed that the random variables y_i,i=1,…,n ,are samples from the probability distribution F which is belong to a parametric distributions class. However, in practice, a parametric model may be inappropriate to describe the data. In this settings, the parametric assumption could be r...
متن کاملConstancy of distributions: asymptotic efficiency of certain nonparametric tests of constancy
In this paper we study stochastic processes which enable monitoring the possible changes of probability distributions over time. These so-called monitoring processes are bivariate functions of time and position at the measurement scale, and in particular be used to test the null hypothesis of no change: one may then form Kolmogorov–Smirnov or other type of tests as functionals of the processes....
متن کاملA risk adjusted self-starting Bernoulli CUSUM control chart with dynamic probability control limits
Usually, in monitoring schemes the nominal value of the process parameter is assumed known. However, this assumption is violated owing to costly sampling and lack of data particularly in healthcare systems. On the other hand, applying a fixed control limit for the risk-adjusted Bernoulli chart causes to a variable in-control average run length performance for patient populations with dissimilar...
متن کاملA Nonparametric Approach to Derivative Asset Pricing
In this research we examine a new method for pricing European call options based on a nonparametric estimate of the probability density of the underlying asset’s returns. Such an approach allows the use of asymmetric and leptokurtic distributions. We estimate the density using a kernel estimation technique applied to random samples drawn from three particular underlying distributions: the Gauss...
متن کامل