The k Nearest Neighbors Estimator of the M-Regression in Functional Statistics

It is well known that the nonparametric estimation of the regression function is highly sensitive to the presence of even a small proportion of outliers in the data.To solve the problem of typical observations when the covariates of the nonparametric component are functional,the robust estimates for the regression parameter and regression operator are introduced.The main propose of the paper is to consider data-driven methods of selecting the number of neighbors in order to make the proposed processes fully automatic.We use thek Nearest Neighbors procedure(kNN)to construct the kernel estimator of the proposed robust model.Under some regularity conditions,we state consistency results for kNN functional estimators,which are uniform in the number of neighbors(UINN).Furthermore,a simulation study and an empirical application to a real data analysis of octane gasoline predictions are carried out to illustrate the higher predictive performances and the usefulness of the kNN approach.

Functional Nonparametric Predictions in Food Industry Using Near-Infrared Spectroscopy Measurement

Functional statistics is a new technique for dealing with data thatcan be viewed as curves or images. Parallel to this approach, the Near-InfraredReflectance (NIR) spectroscopymethodology has been used in modern chemistryas a rapid, low-cost, and exact means of assessing an object’s chemicalproperties. In this research, we investigate the quality of corn and cookiedough by analyzing the spectroscopic technique using certain cutting-edgestatistical models. By analyzing spectral data and applying functional modelsto it, we could predict the chemical components of corn and cookie dough.Kernel Functional Classical Estimation (KFCE), Kernel Functional QuantileEstimation (KFQE), Kernel Functional Expectile Estimation (KFEE),Semi-Partial Linear Functional Classical Estimation (SPLFCE), Semi-PartialLinear Functional Quantile Estimation (SPLFQE), and Semi-Partial LinearFunctional Expectile Estimation (SPLFEE) are models used to accuratelyestimate the different quantities present in Corn and Cookie dough. Theselection of these functional models is based on their ability to constructa forecast region with a high level of confidence. We demonstrate that theconsidered models outperform traditional models such as the partial leastsquaresregression and the principal component regression in terms of predictionaccuracy. Furthermore, we show that the proposed models are morerobust than competing models such as SPLFQE and SPLFEE in the sensethat data heterogeneity has no effect on their efficiency.

Appropriateness of Reduced Modified Three-Parameter Weibull Distribution Function for Predicting Gold Production in Ghana

Forecasting mine production is pertinent to gold mining as it serves as production goals for investors. It is therefore important to identify the exact distribution that gold production as a response variable naturally follows. It is even more appropriate to have a model(s) with few predictor variables. This paper seeks to identify appropriate statistical distribution functions for fitting gold production in Ghana. The empirical paper relied mainly on quarterly secondary datasets on gold production between the years 2009 and 2022 secured from the Minerals Commission of Ghana, Accra. Several known statistical distributions including Weibull, Log-Normal, Generalized Extreme Value (GEV) were explored with Maximum Likelihood Estimation (MLE) and evaluated using model selection criteria as AIC, AICc and BIC. Goodness of Fits were evaluated using Kolmogorov-Smirnov Test (K-S), Cramer-Von Mises Statistic and Anderson-Darling Statistic. Based on the analysis conducted, the reduced modified 3-parameter Weibull distribution provided the best fit for gold production in Ghana. Though the reduced modified Weibull function is proposed, it is important however to recognize that other external factors can influence production levels. Also, the average quarterly fitted gold production is 1000334.8918 ± 75,327.080 (±7.5%) [i.e., 925,007.812 – 1,075,661.972]. This indicates that the average annually fitted gold production lies between 3700031.248 and 4302647.888 ounces at 99.9% confidence level. Therefore, the predicted gold production for the year 2022 is 3.7million ounces at 99.9% confidence level.

Appropriateness of Reduced Modified Three-Parameter Weibull Distribution Function for Predicting Gold Production in Ghana

Forecasting mine production is pertinent to gold mining as it serves as production goals for investors. It is therefore important to identify the exact distribution that gold production as a response variable naturally follows. It is even more appropriate to have a model(s) with few predictor variables. This paper seeks to identify appropriate statistical distribution functions for fitting gold production in Ghana. The empirical paper relied mainly on quarterly secondary datasets on gold production between the years 2009 and 2022 secured from the Minerals Commission of Ghana, Accra. Several known statistical distributions including Weibull, Log-Normal, Generalized Extreme Value (GEV) were explored with Maximum Likelihood Estimation (MLE) and evaluated using model selection criteria as AIC, AICc and BIC. Goodness of Fits were evaluated using Kolmogorov-Smirnov Test (K-S), Cramer-Von Mises Statistic and Anderson-Darling Statistic. Based on the analysis conducted, the reduced modified 3-parameter Weibull distribution provided the best fit for gold production in Ghana. Though the reduced modified Weibull function is proposed, it is important however to recognize that other external factors can influence production levels. Also, the average quarterly fitted gold production is 1000334.8918 ± 75,327.080 (±7.5%) [i.e., 925,007.812 – 1,075,661.972]. This indicates that the average annually fitted gold production lies between 3700031.248 and 4302647.888 ounces at 99.9% confidence level. Therefore, the predicted gold production for the year 2022 is 3.7million ounces at 99.9% confidence level.

New Probability Distributions in Astrophysics: XIII. Truncation for the Benini Distribution

In order to introduce a right truncated version of the Benini distribution, we derive its probability density function, its distribution function, its average value, its kth moment about the origin, its median, how to randomly generate its values, and the maximum likelihood estimator for its three unknown parameters. The astrophysical application of the Benini distribution and its right truncated version is to the initial mass function for stars.

New Probability Distributions in Astrophysics: XII. Truncation for the Gompertz Distribution

Analytical functions which fit the probability distributions of stars and galaxies can provide insight into how these distributions originate. In order to introduce a truncated version of the Gompertz distribution, we derive its probability density function, its distribution function, its average value, its second moment about the origin, its median, its random generation of values and a maximum likelihood estimator for its two unknown parameters. The astrophysical applications of the Gompertz distribution are the initial mass function for stars, the luminosity function for the galaxies of the Sloan Digital Sky Survey, the photometric maximum of galaxies visible in the GLADE+ catalog and a model for the mean absolute magnitude in the GLADE+ catalog as a function of the redshift.

Photon statistics of pulse-pumped four-wave mixing in fiber with weak signal injection

We study the photon statistics of pulse-pumped four-wave mixing in fibers with weak coherent signal injection by measuring the intensity correlation functions of individual signal and idler fields. The experimental results show that the intensity correlation function of individual signal（idler） field g_（s（i））^（2） decreases with the intensity of signal injection. After applying narrow band filter in signal（idler） band, the value of g_（s（i））^（2） decreases from 1.9 ± 0.02（1.9 ± 0.02） to 1.03 ± 0.02（1.05 ± 0.02） when the intensity of signal injection varies from 0 to 120 photons/pulse. The results indicate that the photon statistics changes from Bose–Einstein distribution to Poisson distribution. We calculate the intensity correlation functions by using the multi-mode theory of four-wave mixing in fibers. The theoretical curves well fit the experimental results.Our investigation will be useful for mitigating the crosstalk between quantum and classical channels in a dense wavelength division multiplexing network.

New Probability Distributions in Astrophysics: XI. Left Truncation for the Topp-Leone Distribution

The Topp-Leone (T-L) distribution has aided the modeling of scientific data in many contexts. We demonstrate how it can be adapted to model astrophysical data. We analyse the left truncated version of the T-L distribution, deriving its probability density function (PDF), distribution function, average value, rth moment about the origin, median, the random generation of its values, and its maximum likelihood estimator, which allows us to derive the two unknown parameters. The T-L distribution, in its regular and truncated versions, is then applied to model the initial mass function for the stars. A comparison is made with specific clusters and between proposed functions for the IMF. The Topp-Leone distribution can provide an excellent fit in some cases.

Nonequilibrium Statistical Nature of Intergranular Brittle Fracture in Polycrystal

The brittle fracture probability and reliability are obtained in terms of dislocation mechanism of microcrack evolution. The statistical distribution functions and statistical deviations of elongation, strength, plastic work, crack extension force, fracture foughness, critical and crack length, can be derived in a unified fashion.

A statistical distribution of quad-pol X-band sea clutter time series acquired at a grazing angle

Although the complex Wishart distribution has been widely used to analyze the statistic properties of quad-pol SAR spatial data, the applicability of this complex distribution to the time series of sea clutter is rarely discussed.The measured data of the quad-pol X-band marine radar demonstrate that the time series of the sea echoes are also satisfied the circular Gaussian distributions if the low intensity signals, which are mainly dominated by a radar noise, in the shadow regions of the large-scale waves are removed. On the basis of this fact, the probability density functions（PDFs） of the intensity as well as the phase, the real and the imaginary parts of the sea echoes obtained by the marine radar have been derived, and the theoretical models are all expressed in closed forms. In order to validate the theoretical results, the PDFs are compared with the experimental data collected by the Mc Master IPIX radar. And the comparisons show that the PDF models are in good agreement with the experimental data.

Wave Energy Estimation by Using A Statistical Analysis and Wave Buoy Data near the Southern Caspian Sea

Statistical analysis was done on simultaneous wave and wind using data recorded by discus-shape wave buoy. The area is located in the southern Caspian Sea near the Anzali Port. Recorded wave data were obtained through directional spectrum wave analysis. Recorded wind direction and wind speed were obtained through the related time series as well. For 12-month measurements（May 25 2007-2008）, statistical calculations were done to specify the value of nonlinear auto-correlation of wave and wind using the probability distribution function of wave characteristics and statistical analysis in various time periods. The paper also presents and analyzes the amount of wave energy for the area mentioned on the basis of available database. Analyses showed a suitable comparison between the amounts of wave energy in different seasons. As a result, the best period for the largest amount of wave energy was known. Results showed that in the research period, the mean wave and wind auto correlation were about three hours. Among the probability distribution functions, i.e Weibull, Normal, Lognormal and Rayleigh, ＂Weibull＂ had the best consistency with experimental distribution function shown in different diagrams for each season. Results also showed that the mean wave energy in the research period was about 49.88 k W/m and the maximum density of wave energy was found in February and March, 2010.

Estimating Age in Short Utterances Based on Multi-Class Classification Approach

Age estimation in short speech utterances finds many applications in daily life like human-robot interaction,custom call routing,targeted marketing,user-profiling,etc.Despite the comprehensive studies carried out to extract descriptive features,the estimation errors(i.e.years)are still high.In this study,an automatic system is proposed to estimate age in short speech utterances without depending on the text as well as the speaker.Firstly,four groups of features are extracted from each utterance frame using hybrid techniques and methods.After that,10 statistical functionals are measured for each extracted feature dimension.Then,the extracted feature dimensions are normalized and reduced using the Quantile method and the Linear Discriminant Analysis(LDA)method,respectively.Finally,the speaker’s age is estimated based on a multi-class classification approach by using the Extreme Gradient Boosting(XGBoost)classifier.Experiments have been carried out on the TIMIT dataset to measure the performance of the proposed system.The Mean Absolute Error(MAE)of the suggested system is 4.68 years,and 4.98 years,the Root Mean Square Error(RMSE)is 8.05 and 6.97,respectively,for female and male speakers.The results show a clear relative improvement in terms of MAE up to 28%and 10%for female and male speakers,respectively,in comparison to related works that utilized the TIMIT dataset.

EDGEWORTH EXPANSION FOR CIRCULAR DISTRIBUTION

The paper discusses the Edgeworth expansion for the mean direction =μ(Fn) of directional data and its 'Studentization'. As an application of the results,we give the corresponding result of von Mises population.

Numerical Prediction of Statistical Masonry Wall Fragment Distribution Induced by Blast Loading

The hazard caused by the fragments of damaged structures is usually significant in acci-dental explosions or hostile blast events.A reliable and efficient method to estimate probable fragment size,velocity and launch distance will be useful to assess and design countermeasures to mitigate the possible fragment hazards.This paper presents a numerical method for predicting the size and launch distance of the fragments caused by explosive damage of masonry wall.Numerical simulations with different scaled distances are carried out,and the statistical distribution functions of the fragment size and launch distance in terms of the scaled distance are derived.

New Probability Distributions in Astrophysics: X. Truncation and Mass-Luminosity Relationship for the Frèchet Distribution

The Frèchet distribution has aided the modelling of scientific data in many contexts. We demonstrate how it can be adapted to model astrophysical data. We analyze the truncated version of the Frèchet distribution deriving the probability density function (PDF), the distribution function, the average value, the rth moment about the origin, the median, the random generation of values and the maximum likelihood estimator, which allows us to derive the two unknown parameters. This first PDF in the regular and truncated version is then applied to model the mass of the stars. A canonical transformation from the mass to the luminosity allows us to derive a new PDF, which is derived in its regular and truncated version. Finally, we apply this new PDF model on the distribution in luminosity of NGC 2362.

Estimation of maximum inclusion by statistics of extreme values method in bearing steel

A statistic method, statistics of extreme values （SEV）, was described in detail, which can esti mate the size of maximum inclusion in steel. The characteristic size of the maximum inclusion in a high clean bearing steel （GCrl5） was evaluated by this method, and the morphology and corn position of large inclusions found were analyzed by scanning electron microscopy （SEM）. When standard inspection area （S0） is 280 mm2, the characteristic size of the biggest inclusion found in 30 standard inspection area is 23.93 μm, and it has a 99.9% probability of the characteristic size of maximum inclusion predicted being no larger than 36.85μm in the experimental steel. SEM result shows that large inclusions found are mainly composed of CaS, calcium-aluminate and MgO. Compositing widely exists in large inclusions in high clean bearing steel. Compared with traditional evaluation method, SEV method mainly focuses on inclusion size, and the esti- mation result is not affected by inclusion types. SEV method is suitable for the inclusion eval uation of high clean bearing steel.

THE RAMANUJAN'S Q-EXTENSION FOR THE EXPONENTIAL FUNCTION AND STATISTICAL DISTRIBUTIONS

In this paper. we shall propose a q-extension for the exponential function and develop q-analogs for families of statistical distribution, such as,the normal, and Poisson distribution etc.Many properties of these families will be studied.

Measuring nonlinearity by means of static parameters in Bernoulli binary sequences distribution:A brief approach

This paper analyzes Bernoulli’s binary sequences in the representation of empirical nonlinear events,analyzing the distribution of natural resources,population sizes and other variables that influence the possible outcomes of resource’s usage.Consider the event as a nonlinear system and the metrics of analysis consisting of two dependent random variables 0 and 1,with memory and probabilities in maximum finite or infinite lengths,constant and equal to 1/2 for both variables(stationary process).The expressions of the possible trajectories of metric space represented by each binary parameter remain constant in sequences that are repeated alternating the presence or absence of one of the binary variables at each iteration(symmetric or asymmetric).It was observed that the binary variables X_(1)and X_(2)assume on time T_(k)→∞specific behaviors(geometric variable)that can be used as management tools in discrete and continuous nonlinear systems aiming at the optimization of resource’s usage,nonlinearity analysis and probabilistic distribution of trajectories occurring about random events.In this way,the paper presents a model of detecting fixed-point attractions and its probabilistic distributions at a given population-resource dynamic.This means that coupling oscillations in the event occur when the binary variables X_(1)and X_(2)are limited as a function of time Y.