JournalTOCs

Annals of Data Science
Number of Followers: 17

Hybrid journal (It can contain Open Access articles)
ISSN (Print) 2198-5804 - ISSN (Online) 2198-5812
Published by Springer-Verlag

[2467 journals]

A Statistical Analysis of Chinese Stock Indices Returns From Approach of
Parametric Distributions Fitting
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The stock price process in China is full of uncertainty hence the stock indices were introduced to serve as indicators of the financial market. How to capture the statistical characteristics of Chinese stock indices returns by the method of parametric distributions fitting could be useful in the fields of econometrics and risk management. In this paper, we use a wider range of parametric distributions to model four main Chinese stock indices. We find a generalization of the Student’s t distribution is shown to provide the best fit.
  PubDate: 2023-02-01
Sparse Principal Component Analysis for Natural Language Processing
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract High dimensional data are rapidly growing in many different disciplines, particularly in natural language processing. The analysis of natural language processing requires working with high dimensional matrices of word embeddings obtained from text data. Those matrices are often sparse in the sense that they contain many zero elements. Sparse principal component analysis is an advanced mathematical tool for the analysis of high dimensional data. In this paper, we study and apply the sparse principal component analysis for natural language processing, which can effectively handle large sparse matrices. We study several formulations for sparse principal component analysis, together with algorithms for implementing those formulations. Our work is motivated and illustrated by a real text dataset. We find that the sparse principal component analysis performs as good as the ordinary principal component analysis in terms of accuracy and precision, while it shows two major advantages: faster calculations and easier interpretation of the principal components. These advantages are very helpful especially in big data situations.
  PubDate: 2023-02-01
Progressive Type-II Censored Data and Associated Inference with
Application Based on Li–Li Rayleigh Distribution
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Based on progressive Type-II censored samples, we first derive the recurrence relations for the single and product moments of progressively Type-II censored order statistics from two parameter Rayleigh distribution. These recurrence relations enable us to compute the mean and variances of all progressively Type-II censored order statistics for all sample sizes in a simple and efficient manner. Further, an algorithm is discussed which enable us to compute all the means and variances of two parameter Rayleigh progressive Type-II censored order statistics for all sample sizes and all censoring schemes. Next, we obtain the maximum likelihood estimators of the unknown parameters and the approximate confidence intervals of the parameters of the Rayleigh distribution. Finally, we consider Bayes estimation under five different types of loss functions (symmetric and asymmetric loss functions) using independent gamma priors for both the unknown parameters. Monte Carlo simulations are performed to compare the performance of the proposed methods, and one data set has been analyzed for illustrative purposes.
  PubDate: 2023-02-01
A Study on Academic Attainment of Agriculture Students and its Correlates:
A Dummy Regression Approach
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Education is a Nation’s strength. Association analysis of academic performance and its influential factors has remained research interest for all education researchers all over the world. India being an agriculture dominated country, for its development in agricultural front it requires ahuge numberof efficient technocrats having strong academic background. In this study an attempt has been made to examine the associationship of academic performance of the agriculture graduates, as measured through overall grade point average (OGPA) with the factors supposed to influence the academic performance. Special emphasis has been given to visualize the performance in presence of the influences of nominal factors. Students at masters level were surveyed for their social, economic, demographic and family and educational background through a designed questionnaire and tested accordingly. Statistical tools, starting from frequency, percentage, Chi-square test, test for normality, Cramer’s V test, multiple regression analysis with the inclusion of dummy variables were employed. Dependency of OGPA with gender, caste and expenditure on education is recorded. The dependency of educational expenditure on OGPA is quite obvious. But the dependency of OGPA with those of gender and caste is most probably not a good sign for a healthy higher education system. This study will help the education planners to take group oriented action plan for improving the education standard in higher education institutions.
  PubDate: 2023-02-01
A Comprehensive Comparative Study of Artificial Neural Network (ANN) and
Support Vector Machines (SVM) on Stock Forecasting
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract From exchanging budgetary instruments to tracking individual spending plans to detail a business's profit, money-related organisations utilise computational innovation day by day. Here in this paper, we focus on the significance of innovation in accounts such as financial risk management and stock prediction. We discuss two significant algorithms that have a notable role in stock forecasting. Artificial Neural Networks (ANN), as absenteeism of some data points, does not hamper the network functioning. Secondly, Support Vector Machines (SVM) has several features, and due to simple decision boundaries, it avoids over-fitting. The paper first looks at the different technologies applied in stock market prediction. It examines how sentimental analysis, decision trees, moving average algorithm, and data mining is applied in various stock prediction scenarios. The paper covers the recent past studies to explore the concepts and methodologies through which ANN's and SVM's have been used. Additionally, the paper incorporates significant aspects of novel methods and technologies in which ANN as a hybrid model like ANN-MLP, GARCH-MLP, a combination of the Backpropagation algorithm and Multilayer Feed-forward network, yields better results. Simultaneously, SVM's have been successfully applied in stock prediction, giving an accuracy of about 60%–70% for simple SVM, which is further improved by combining methods like Random Forest, Genetic Algorithm more accurate outcomes. Further, we present our thoughts on where SVM's and ANN's stand as prediction algorithms and challenges like the time constraint, current scenarios, data limitation, and cold start problems were raised. Conclusively SVM and ANN played prominent roles in tackling these issue to an extent and can further be enhanced with their integration with other novel techniques resulting in hybrid methodologies. It will lead students, researchers and financial enthusiasts to more potent approaches for Stock forecasting.
  PubDate: 2023-02-01
Application of Generalized Regression Neural Network in Predicting the
Performance of Solar Photovoltaic Thermal Water Collector
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Solar photovoltaic thermal water collector (SPV/T-WC) is a hybrid device which converts power from the solar energy in to thermal and electrical simultaneously. The performance of such SPV/T-WC mainly depends on its electrical and thermal power output. Besides the performance of SPV/T-WC, is more sensitive to the transient nature of electrical and thermal power output. Thus a demand for predicting the performance variations in the SPV/T-WC is demand by users. Only limited performance prediction based research works are attempted in the performance prediction of the SPV/T-WC either numerically or by using cognitive models. In this study, two generalized regression neural network (GRNN) models are proposed to predict the transient performance variations in the SPV/T-WC. The two individual objectives of the first and second model include the prediction of overall power output and the overall efficiency delivered by an SPV/T-WC system. Both the GRNN models proposed in this study consist of two inputs and single output. In order to train this GRNN model, real time experiments are conducted with stand-alone SPV/T-WC for four continuous days. Then based on such experimental data sets, GRNN models are trained, tested, and validated. The results predicted by the both GRNN models are in good agreement with the real time experimental results. The overall accuracy of the proposed GRNN models in predicting the performance is 95.36% and 96.22% respectively.
  PubDate: 2023-02-01
Estimation of Domain Mean Using Conventional Synthetic Estimator with Two
Auxiliary Characters
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The estimation of domain mean is being accelerated applied to draft program policy in the government and private sectors. The use of two auxiliary characters is better choice as compared to single auxiliary character. The main interest is to consist information about an additional auxiliary character z in auxiliary character x and utilize for interested domain. This paper has investigated conventional generalized synthetic estimator for domain mean using two auxiliary characters x and z, and also discussed its properties. A comparative study of the proposed estimator has been made with the conventional ratio and conventional generalized estimators in terms of absolute relative bias and simulated relative standard error. It has evaluated, the proposed estimator is more efficient than the relevant estimators.
  PubDate: 2023-02-01
Bayesian Effective Biological Dose Determination in Immunotherapy Response
Trial
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Immunotherapy, especially checkpoint inhibitors, have transformed the treatment of cancer. Unlike chemotherapy, checkpoint inhibitors modify and enable the patient's immune system to fight cancer, thus prolonging survival. The conventional maximum tolerable dose finding designs were used for dose-finding in checkpoint inhibitors studies. These proved to be unsuitable as in the majority of checkpoint inhibitors there was no appearance of toxicity. Hence doses were selected using pharmacokinetic and pharmacodynamic modelling. However, these doses produce plasma levels of the drug, which are far higher than the levels required for its optimal action. Further increment in dose in phase 1 settings was not associated with an increment in response or survival. Considering the cost implications and scarcity of these resources probably a dose much higher than necessary is administered. The need of the hour is to define a dose beyond which in the majority of patients, there won't be an incremental benefit in cancer-related outcomes. The current challenge is that to best of our knowledge, and no statistical model exists to find the minimally effective dose of the checkpoint inhibitors. Therefore, here we propose a Bayesian design to determine the effective biological dose (EBD) for immunotherapy trials. This work is about the preparation of methodology with two scenarios, (1) EBD of checkpoint inhibitors administered as monotherapy (2) EBD of checkpoint inhibitors administered as a combined therapy.
  PubDate: 2023-02-01
Estimating Reliability Characteristics of the Log-Logistic Distribution
Under Progressive Censoring with Two Applications
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Let a progressively type-II (PT-II) censored sample of size m is available. Under this set-up, we consider the problem of estimating unknown model parameters and two reliability characteristics of the log-logistic distribution. Maximum likelihood estimates (MLEs) are obtained. We use expectation–maximization (EM) algorithm. The observed Fisher information matrix is computed. We propose Bayes estimates with respect to various loss functions. In this purpose, we adopt Lindley’s approximation and importance sampling methods. Asymptotic and bootstrap confidence intervals are derived. Asymptotic intervals are obtained using two approaches: normal approximation to MLEs and log-transformed MLEs. The bootstrap intervals are computed using boot-t and boot-p algorithms. Further, highest posterior density (HPD) credible intervals are constructed. Two sets of practical data are analyzed for the illustration purpose. Finally, detailed simulation study is carried out to observe the performance of the proposed methods.
  PubDate: 2023-02-01
A New Extension of the Topp–Leone-Family of Models with Applications
to Real Data
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract In this article, we proposed a new extension of the Topp–Leone family of distributions. Some important properties of the model are developed, such as quantile function, stochastic ordering, model series representation, moments, stress–strength reliability parameter, Renyi entropy, order statistics, and moment of residual life. A particular member called new extended Topp–Leone exponential (NETLE) is discussed. Maximum likelihood estimation (MLE), least-square estimation (LSE), and percentile estimation (PE) are used for the model parameter estimation. Simulation studies were conducted using NETLE to assess the MLE, LSE, and PE performance by examining their bias and mean square error (MSE), and the result was satisfactory. Finally, the applications of the NETLE to two real data sets are provided to illustrate the importance of the NETLG families in practice; the data sets consist of daily new deaths due to COVID-19 in California and New Jersey, USA. The new model outperformed many other existing Topp–Leone’s and exponential related distributions based on the real data illustrations.
  PubDate: 2023-02-01
Comparison Between Dependent and Independent Ranked Set Sampling Designs
for Parametric Estimation with Applications
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract This paper is concerned with the estimation problem using maximum likelihood method of estimation for the unknown parameters of exponetiated gumbel distribution based on neoteric ranked set sampling (NRSS) as a new modification of the usual ranked set sampling (RSS) technique. Numerical study is conducted to compare NRSS as a dependent ranked set sampling technique, with RSS, and median ranked set sampling as independent sampling techniques, and then the performance of RSS and its modifications will be compared with simple random sampling based on their mean square errors and efficiencies.
  PubDate: 2023-02-01
Inferences Based on Correlated Randomly Censored Gumbel’s Type-I
Bivariate Exponential Distribution
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The formal random censoring plan has been extensively studied earlier in statistical literature by numerous researchers to deal with dropouts or unintentional random removals in life-testing experiments. All of them considered failure time and censoring time to be independent. But there are several situations in which one observes that as the failure time of an item increases, the censoring time decreases. In medical studies or especially in clinical trials, the occurrence of dropouts or unintentional removals is frequently observed in such a way that as the treatment (failure) time increases, the dropout (censoring) time decreases. No work has yet been found that deals with such correlated failure and censoring times. Therefore, in this article, we assume that the failure time is negatively correlated with censoring time, and they follow Gumbel’s type-I bivariate exponential distribution. We compute the maximum likelihood estimates of the model parameters. Using the Monte Carlo Markov chain methods, the Bayesian estimators of the parameters are calculated. The expected experimental time is also evaluated. Finally, for illustrative purposes, a numerical study and a real data set analysis are given.
  PubDate: 2023-01-31
Bayesian Hierarchical Spatial Modeling of COVID-19 Cases in Bangladesh
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract This research aimed to investigate the spatial autocorrelation and heterogeneity throughout Bangladesh’s 64 districts. Moran I and Geary C are used to measure spatial autocorrelation. Different conventional models, such as Poisson-Gamma and Poisson-Lognormal, and spatial models, such as Conditional Autoregressive (CAR) Model, Convolution Model, and modified CAR Model, have been employed to detect the spatial heterogeneity. Bayesian hierarchical methods via Gibbs sampling are used to implement these models. The best model is selected using the Deviance Information Criterion. Results revealed Dhaka has the highest relative risk due to the city’s high population density and growth rate. This study identifies which district has the highest relative risk and which districts adjacent to that district also have a high risk, which allows for the appropriate actions to be taken by the government agencies and communities to mitigate the risk effect.
  PubDate: 2023-01-22
Intervention Analysis of COVID-19 Vaccination in Nigeria: The Naive
Solution Versus Interrupted Time Series
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract In this paper, an intervention analysis approach was applied to daily cases of COVID-19 in Nigeria in order to evaluate the utilization and effect of the COVID-19 vaccine administered in the country. Data on the daily report of COVID-19 cases in Nigeria were collected and subjected to two models: the naïve solution and the interrupted time series (the intervention model). Based on the Alkaike Information Criterion (AIC), sigma2, and log likelihood values, the interrupted time series model outperformed the Naïve solution model. ARIMA (4, 1, 4) with exogenous variables was identified as the best model. It was observed that the intervention (vaccination) was not significant at the 5% level of significance in reducing the number of daily COVID-19 cases in Nigeria since the start of the vaccination on March 5, 2021, until March 28, 2022. Also, the ARIMA (4, 1, 4) forecasts indicated that there will be surge in the number of daily COVID-19 cases in Nigeria between January and April 2023. As a result, we recommend strict adherence to COVID-19 protocols as well as further vaccination and sensitization programs to educate people on the importance of vaccine uptake and avoid Corona virus spread in the year 2023 and beyond.
  PubDate: 2023-01-19
Exchange Rate Forecasting: Nonlinear GARCH-NN Modeling Approach
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract This paper targets the description of the fusion of modeling techniques, such as the GARCH model and the Artificial Neural Network (ANN), for the sake of predicting financial series and precisely the series of the exchange rate in Tunisia, namely the USD/TND, the EUR/TND and the YEN/TND, for a daily frequency extending from 2015 through 2019. To our knowledge, this is the only paper that focuses on the descriptions of the fusion of modeling techniques (GARCH-NN) or hybridization method that applied on Tunisian currency (TND). The empirical results show that the hybrid model (GARCH-NN) outperforms and it is more efficient than the two used models. In fact, this method combines the advantages of two approaches in order to obtain a result more satisfactory for the expectations of the policy-makers in the exchange market in order to take their decisions. The results showed that the model proposed gives better results, so, can be an alternative of standard linear autoregressive model. This result has been joined by many empirical studies that confirm the quality and performance of this methodology, which researchers advise to be retained in all areas.
  PubDate: 2023-01-03
An Alternative to the Beta Regression Model with Applications to OECD
Employment and Cancer Data
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract In regression analysis involving response variable on the bounded unit interval [0, 1], the beta regression model often suffice as a common choice, however, there are many alternatives to the beta regression model. In this article, we add yet another new alternative to the literature called the unit upper truncated Weibull (unit UTW) regression model. We introduce a novel unit UTW distribution as an alternative to the beta distribution and we present some of its mathematical properties. The unit UTW distribution is then extended to build the unit UTW regression model. Through an extensive Monte-Carlo simulation experiments, we show that the method of maximum likelihood can provide good estimate for each parameter in the new models. We give two practical examples were the proposed models performed better than the beta distribution and the beta regression model.
  PubDate: 2022-12-27
Correction to: Guest Editor’s Introduction: COVID-19 and Data
Science
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  PubDate: 2022-12-01
  DOI: 10.1007/s40745-022-00447-z
Application of New Companding Techniques on the DWT-Based SC-FDMA System
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Orthogonal Frequency Division Multiplexing (OFDM) is the most popular multicarrier communication technique, but its disadvantage is the large Peak-to-Average Power Ratio (PAPR). In recent years, different researchers presented several techniques to avoid this problem such as companding techniques. Moreover, the Single-Carrier Frequency Division Multiple Access (SC-FDMA) system is a popular system in mobile communications because of its advantage of low PAPR, but reducing its PAPR is still an open research issue. So, an extension of the work applied on OFDM to SC-FDMA is adopted in this paper to reduce the PAPR, while achieving a low Bit Error Rate (BER). New companding schemes are adopted in this paper with the help of the Discrete Wavelet Transform (DWT) in the presence of channel degradations for lowering the PAPR of the SC-FDMA system, while achieving a low BER.
  PubDate: 2022-12-01
  DOI: 10.1007/s40745-022-00413-9
Offline Signature Verification: An Application of GLCM Features in Machine
Learning
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Signatures are a crucial behavioral trait widely used to authenticate a person's identity. Financial and legal institutions, including commercial banks, consider it a legitimate method of document authentication. Despite the emergence of various biometric authentication techniques such as fingerprints, retinal scans, and facial recognition, signature verification is still a prevalent authentication method among Indian industries, especially in the banking sector. Signature verification is used while processing cheques and other documents, even when only digital copies of such documents are available. An example of signature verification on digital documents is the Cheque Truncation System of India, adopted by all scheduled commercial banks in India. However, manual signature verification is tedious and vulnerable to human error. This paper attempts to compare the efficacy of Convolution Neural Networks and Support Vector Machine algorithms in automating the process of signature verification. These algorithms incorporate various image features to verify whether the signature is genuine or fraudulent without human intervention. The Support Vector Machine algorithm performs better, considering the computational limitations of production systems.
  PubDate: 2022-12-01
  DOI: 10.1007/s40745-021-00343-y
Human Capital in Public Research Laboratory: Towards an Alternative
Evaluation and Prediction Method Based on Hybridization
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The alternative proposed method aims to combine management accounting, precisely the activity-based system, and statistical tools to develop a method of assessing and predicting human capital within research laboratory. Statistical tools are Standardized Mean Difference, Hierarchical Cluster Analysis and Discriminant Analysis. The first normalizes the activities of the laboratory; the second classifies the results obtained, while the third standardizes these results by expressing them in terms of probability. The standardized scores are used for the computation and the prediction of human capital in research laboratories via activity regrouping center. The originality of this work is to fill a research gap in the field of hybridization in calculation and prediction of human capital by integrating the two disciplines mentioned above. Likewise, the originality of this work lies in the use of an activity-based accounting architecture to process outputs (and not costs) related to intangible aspects. The proposed method has research and social implications since it allows making appropriate research policy, adequate management control and improves organizational relations within the laboratory concerned. The findings show, through an illustration, the applicability of the proposed method and the usefulness of the tools used to do this.
  PubDate: 2022-12-01
  DOI: 10.1007/s40745-020-00241-9