检索结果-南通市图书馆

Limit Distribution of the φ-Diverg.nce Based Chang. Point Estimator

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2021年第3期11卷 337-350页

作者： Mwelu Susan anthony g. waititu Peter N. Mwita Charity Wamwea Pan African University Institute for Basic Sciences Technolog. and Innovation Nairobi Kenya Department of Statistics and Actuarial Sciences JKUAT Nairobi Kenya Department of Mathematics Statistics and Actuarial Sciences Machakos University Machakos Kenya

The assumption of stationarity is too restrictive especially for long.time series. This paper studies the chang. point problem throug. a chang. point estimator based on the φ-divergence which provides a rich set of distance like measures between pairs of distributions. The change point problem is considered in the following sub-fields: the problem of divergence estimation, testing for the homogeneity between two samples as well as estimating the time of change. The asymptotic distribution of the change point estimator is estimated by the limiting distribution of a stochastic process within given bounds through asymptotic theory surrounding the likelihood theory. The distribution is found to converge to that of a standardized Brownian bridge process.

关键词： Chang. Point φ-Diverg.nce Estimator Test Statistic Distribution Wald

Spatial Heterog.neity Modeling.Using.Machine Learning.Based on a Hybrid of Random Forest and Convolutional Neural Network (CNN)

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Data Analysis and Information Processing 2024年第3期12卷 319-347页

作者： Amadou Kindy Barry anthony waititu g.chuhi Lawrence Nderu Department of Mathematics Pan African University Institute for Basic Sciences Technology and Innovation (PAUISTI) Nairobi Kenya Department of Statistics and Actuarial Science Jomo Kenyatta University of Agriculture and Technology Nairobi Kenya Department of Computing.School and Information Technolog. (SCIT) Jomo Kenyatta University of Agriculture and Technology (JKUAT) Nairobi Kenya

Spatial heterog.neity refers to the variation or differences in characteristics or features across different locations or areas in space. Spatial data refers to information that explicitly or indirectly belong. to a particular g.og.aphic reg.on or location, also known as g.o-spatial data or g.og.aphic information. Focusing.on spatial heterog.neity, we present a hybrid machine learning.model combining.two competitive alg.rithms: the Random Forest Reg.essor and CNN. The model is fine-tuned using.cross validation for hyper-parameter adjustment and performance evaluation, ensuring.robustness and g.neralization. Our approach integ.ates g.obal Moran’s I for examining.g.obal autocorrelation, and local Moran’s I for assessing.local spatial autocorrelation in the residuals. To validate our approach, we implemented the hybrid model on a real-world dataset and compared its performance with that of the traditional machine learning.models. Results indicate superior performance with an R-squared of 0.90, outperforming.RF 0.84 and CNN 0.74. This study contributed to a detailed understanding.of spatial variations in data considering.the g.og.aphical information (Long.tude & Latitude) present in the dataset. Our results, also assessed using.the Root Mean Squared Error (RMSE), indicated that the hybrid yielded lower errors, showing.a deviation of 53.65% from the RF model and 63.24% from the CNN model. Additionally, the g.obal Moran’s I index was observed to be 0.10. This study underscores that the hybrid was able to predict correctly the house prices both in clusters and in dispersed areas.

关键词： Spatial Heterog.neity Spatial Data Feature Selection Standardization Machine Learning.Models Hybrid Models

Autonomous Surveillance of Infants’ Needs Using.CNN Model for Audio Cry Classification

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Data Analysis and Information Processing 2022年第4期10卷 198-219页

作者： g.ofrey Owino anthony waititu anthony Wanjoya John Okwiri Department of Statistics and Actuarial Sciences Jomo Kenyatta University of Agriculture and Technology Nairobi Kenya

Infants portray sug.estive unique cries while sick, having.belly pain, discomfort, tiredness, attention and desire for a chang. of diapers among.other needs. There exists limited knowledg. in accessing.the infants’ needs as they only relay information throug. sug.estive cries. Many teenag.rs tend to g.ve birth at an early ag., thereby exposing.them to be the key monitors of their own babies. They tend not to have sufficient skills in monitoring.the infant’s dire needs, more so during.the early stag.s of infant development. Artificial intellig.nce has shown promising.efficient predictive analytics from supervised, and unsupervised to reinforcement learning.models. This study, therefore, seeks to develop an android app that could be used to discriminate the infant audio cries by leverag.ng.the streng.h of convolution neural networks as a classifier model. Audio analytics from many kinds of literature is an untapped area by researchers as it’s attributed to messy and hug. data g.neration. This study, therefore, strong.y leverag.s convolution neural networks, a deep learning.model that is capable of handling.more than one-dimensional datasets. To achieve this, the audio data in form of a wave was converted to imag.s throug. Mel spectrum frequencies which were classified using.the computer vision CNN model. The Librosa library was used to convert the audio to Mel spectrum which was then presented as pixels serving.as the input for classifying.the audio classes such as sick, burping. tired, and hung.y. The study g.al was to incorporate the model as an android tool that can be utilized at the domestic level and hospital facilities for surveillance of the infant’s health and social needs status all time round.

关键词： Convolutional Neural Network (CNN) Mel Frequency Cepstral Coefficients (MFCCs) Rectified Linear Unit (ReLU) Activation Function Audio Analytics Deep Neural Network (DNN)

A Hybrid DNN-RBFNN Model for Intrusion Detection System

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Data Analysis and Information Processing 2023年第4期11卷 371-387页

作者： Wafula Maurice Oboya anthony waititu g.chuhi anthony Wanjoya Department of Statistics and Actuarial Sciences Jomo Kenyatta University of Agriculture and Technology Nairobi Kenya

Intrusion Detection Systems (IDS) are pivotal in safeg.arding.computer networks from malicious activities. This study presents a novel approach by proposing.a Hybrid Dense Neural Network-Radial Basis Function Neural Network (DNN-RBFNN) architecture to enhance the accuracy and efficiency of IDS. The hybrid model synerg.zes the streng.hs of both dense learning.and radial basis function networks, aiming.to address the limitations of traditional IDS techniques in classifying.packets that could result in Remote-to-local (R2L), Denial of Service (Dos), and User-to-root (U2R) intrusions.

关键词： Dense Neural Network (DNN) Radial Basis Function Neural Network (RBFNN) Intrusion Detection System (IDS) Denial of Service (DoS) Remote to Local (R2L) User-to-Root (U2R)

A Hybrid Neural Network Model Based on Transfer Learning.for Forecasting.Forex Market

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Data Analysis and Information Processing 2023年第2期11卷 103-120页

作者： Salum Hassan Faru anthony waititu Lawrence Nderu Department of Mathematics The Pan African University Institute for Basic Sciences Technology and Innovation (PAUSTI) Nairobi Kenya Department of Statistics and Actuarial Science Jomo Kenyatta University of Agriculture and Technology Juja Kenya Department of Computing.and Information Technolog. Jomo Kenyatta University of Agriculture and Technology Juja Kenya

The forecasting.research literature has developed g.eatly in recent years as a result of advances in information technolog.. Financial time-series tasks have made substantial use of machine learning.and deep neural networks, but building.a prediction model from scratch takes time and computational resources. Transfer learning.is g.owing.popular in tackling.these constraints of training.time and computational resources in several disciplines. This study proposes a hybrid base model for the financial time series prediction employing.the recurrent neural network (RNN) and long.short term memory (LSTM) called RNN-LSTM. We used random search to fine-tune the hyperparameters and compared our proposed model to the RNN and LSTM base models and evaluate using.the RMSE, MAE, and MAPE metrics. When forecasting.Forex currency pairs g.P/USD, USD/ZAR, and AUD/NZD our proposed base model for transfer learning.outperforms RNN and LSTM base model with root mean squared errors of 0.007656, 0.165250, and 0.001730 respectively.

关键词： Deep Learning Transfer Learning Time Series Analysis RNN LSTM

A Hybrid Spatial Dependence Model Based on Radial Basis Function Neural Networks (RBFNN) and Random Forest (RF)

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Data Analysis and Information Processing 2023年第3期11卷 293-309页

作者： Mamadou Hady Barry Lawrence Nderu anthony waititu g.chuhi Department of Mathematics Pan African University Institute for Basic Sciences Technology and Innovation (PAUSTI) Nairobi Kenya Department of Computing.School and Information Technolog. (SCIT) Jomo Kenyatta University of Agriculture and Technology (JKUAT) Nairobi Kenya Department of Statistics and Actuarial Science Jomo Kenyatta University of Agriculture and Technology Nairobi Kenya

The majority of spatial data reveal some deg.ee of spatial dependence. The term “spatial dependence” refers to the tendency for phenomena to be more similar when they occur close tog.ther than when they occur far apart in space. This property is ig.ored in machine learning.(ML) for spatial domains of application. Most classical machine learning.alg.rithms are g.nerally inappropriate unless modified in some way to account for it. In this study, we proposed an approach that aimed to improve a ML model to detect the dependence without incorporating.any spatial features in the learning.process. To detect this dependence while also improving.performance, a hybrid model was used based on two representative alg.rithms. In addition, cross-validation method was used to make the model stable. Furthermore, g.obal moran’s I and local moran were used to capture the spatial dependence in the residuals. The results show that the HM has sig.ificant with a R2 of 99.91% performance compared to RBFNN and RF that have 74.22% and 82.26% as R2 respectively. With lower errors, the HM was able to achieve an averag. test error of 0.033% and a positive g.obal moran’s of 0.12. We concluded that as the R2 value increases, the models become weaker in terms of capturing.the dependence.

关键词： Spatial Data Spatial Dependence Hybrid Model Machine Learning.Alg.rithms

Bull and Bear Dynamics of the Nig.ria Stock Returns Transitory via Ming.ed Autoreg.essive Random Processes

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2021年第5期11卷 870-885页

作者： Rasaki Olawale Olanrewaju anthony g.chuhi waititu Lukman Abiodun Nafiu Department of Mathematical Sciences Pan African University Institute for Basic Sciences Technology and Innovation Nairobi Kenya Department of Mathematical Sciences Jomo Kenyatta University of Agriculture and Technology Nairobi Kenya Department of Mathematics and Statistics Kabale University Kabale Uganda

This paper expounds the nitty-g.itty of stock returns transitory, periodical behavior of its markets’ demands and cyclical-like tenure-chang.ng.of number of the stocks sold. Ming.ing.of autoreg.essive random processe... 详细信息

This paper expounds the nitty-g.itty of stock returns transitory, periodical behavior of its markets’ demands and cyclical-like tenure-changing of number of the stocks sold. Mingling of autoregressive random processes via Poisson and Extreme-Value-Distributions (Fréchet, Gumbel, and Weibull) error terms were designed, generalized and imitated to capture stylized traits of k-serial tenures (ability to handle cycles), Markov transitional mixing weights, switching of mingling autoregressive processes and full range shape changing predictive distributions (multimodalities) that are usually caused by large fluctuations (outliers) and long-memory in stock returns. The Poisson and Extreme-Value-Distributions Mingled Autoregressive (PMA and EVDs) models were applied to a monthly number of stocks sold in Nigeria from 1960 to 2020. It was deduced that fitted Gumbel-MAR (2:1, 1) outstripped other linear models as well as best fitted among the Poisson and Extreme-Value-Distributions Mingled autoregressive models subjected to the discrete monthly stocks sold series.

关键词： Autoreg.essive Random Processes Extreme-Value-Distributions Ming.ed Poisson Stock Returns

ANN-Time Varying.g.RCH Model: Simulations and Application in Modelling.Temperature for Weather Derivatives

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2022年第3期12卷 433-441页

作者： Elias K. Karuiru John Mwaniki Kihoro Thomas Mag.to anthony g.chuhi waititu The Pan African University Institute for Basic Sciences Technology and Innovation (PAUSTI) Nairobi Kenya School of Computing.and Mathematics Co-Operative University Nairobi Kenya Department of Statistics and Actuarial Sciences JKUAT Nairobi Kenya

In economics and finance, minimising.errors while building.an abstract representation of financial assets plays a critical role due to its application in areas such as risk manag.ment, decision making.and option pricing. Despite the many methods developed to handle this problem, modelling.processes with fixed and random periodicity still remains a major challeng.. Such methods include Artificial Neural networks (ANN), Fuzzy Inference system (FIS), g.RCH models and their hybrids. This study seeks to extend literature of hybrid ANN-Time Varying.g.RCH model throug. simulations and application in modelling.weather derivatives. The study models daily temperature of Kenya using.ANN-Time Varying.g.RCH (1, 1), Time Lag.ed Feedforward neural network (TLNN) and periodic g.RCH family models. Mean square error (MSE) and coefficient of determination R2 were used to determine performance of the models under study. Results obtained show that the ANN-Time Varying.g.RCH model g.ves the best results.

关键词： Artificial Neural Network Time Varying.g.RCH Weather Derivatives Temperature

ANN-Time Varying.g.RCH Model for Processes with Fixed and Random Periodicity

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Open Journal of Statistics 2021年第5期11卷 673-689页

Financial Time Series Forecasting.is an important tool to support both individual and org.nizational decisions. Periodic phenomena are very popular in econometrics. Many models have been built aiding.capture of these periodic trends as a way of enhancing.forecasting.of future events as well as g.iding.business and social activities. The nature of real-world systems is characterized by many uncertain fluctuations which makes prediction difficult. In situations when randomness is mixed with periodicity, prediction is even much harder. We therefore constructed an ANN Time Varying Garch model with both linear and non-linear attributes and specific for processes with fixed and random periodicity. To eliminate the need for time series linear component filtering, we incorporated the use of Artificial Neural Networks (ANN) and constructed Time Varying.g.RCH model on its disturbances. We developed the estimation procedure of the ANN time varying.g.RCH model parameters using.non parametric techniques.

关键词： Fixed Periodicity Random Periodicity Artificial Neural Network Time Varying.g.RCH