Maximum Likelihood Estimation Technique Research Articles

Nonparametric mean and variance adaptive classification rule for high‐dimensional data with heteroscedastic variances

AbstractIn this study, we introduce an innovative methodology aimed at enhancing Fisher's Linear Discriminant Analysis (LDA) in the context of high‐dimensional data classification scenarios, specifically addressing situations where each feature exhibits distinct variances. Our approach leverages Nonparametric Maximum Likelihood Estimation (NPMLE) techniques to estimate both the mean and variance parameters. By accommodating varying variances among features, our proposed method leads to notable improvements in classification performance. In particular, unlike numerous prior studies that assume the distribution of heterogeneous variances follows a right‐skewed inverse gamma distribution, our proposed method demonstrates excellent performance even when the distribution of heterogeneous variances takes on left‐skewed, symmetric, or right‐skewed forms. We conducted a series of rigorous experiments to empirically validate the effectiveness of our approach. The results of these experiments demonstrate that our proposed methodology excels in accurately classifying high‐dimensional data characterized by heterogeneous variances.

Assessing the impact of non-tariff measures on Sri Lankan mango exports: insights, challenges, and recommendations

IntroductionWith the increase in non-tariff measures (NTMs) in recent years, understanding their impact on trade policies and agricultural exports is of utmost importance for countries like Sri Lanka, where the proliferation of NTMs has posed substantial impediments to its mango export industry. However, studying NTMs can be challenging due to their complex nature, diverse range, and limited information availability.MethodsThis research aims to investigate the significant effect of NTMs on Sri Lankan mango exports, identify the pivotal drivers influencing mango exports, and analyze the performance of mango exports in Sri Lanka. This study utilized a panel dataset of 16 importing countries spanning the period from 2000 to 2021. The primary focus was on examining NTM-related challenges and obstacles faced by the mango export industry in Sri Lanka. The identification and quantification of these obstacles were achieved through a multifaceted approach in which assessment of NTM-related rejections, both domestic and overseas, evaluation of NTM incidences using frequency and coverage ratio measures, and the deployment of various gravity model equations to quantify the effect of NTMs were taken into consideration. Probit, Ordinary Least Square, Random Effects, and Poisson Pseudo Maximum Likelihood estimation techniques were utilized for data analysis.ResultsThe findings indicate a significant inverse correlation between the increase in the total count of NTMs and Sri Lankan mango exports if the country aims to export more mangoes to relevant markets, while Sanitary and Phytosanitary Measures (SPS), Technical Barriers to Trade (TBT), and Pre-shipment Inspection (PSI) contributed to increased mango exports. Among other explanatory variables, the GDP of both importing and exporting countries had a positive influence on Sri Lankan mango exports. Surprisingly, the distance to the importing country had no significant effect, though it shows a negative sign.DiscussionThis study provides valuable insights into the impact of NTMs on Sri Lankan mango exports. It highlights the considerable influence of NTMs on the overall fresh fruit export sector. We recommend proactively aligning internal quality testing regulations with the standards required by destination countries to promote future expansion.

Generalization of Odd Ramos-Louzada generated family of distributions: Properties, characterizations, and applications to diabetes and cancer survival datasets

Probability distributions offer the best description of survival data and as a result, various lifetime models have been proposed. However, some of these survival datasets are not followed or sufficiently fitted by the existing proposed probability distributions. This paper presents a novel Kumaraswamy Odd Ramos-Louzada-G (KumORL-G) family of distributions together with its statistical features, including the quantile function, moments, probability-weighted moments, order statistics, and entropy measures. Some relevant characterizations were obtained using the hazard rate function and the ratio of two truncated moments. In light of the proposed KumORL-G family, a five-parameter sub-model, the Kumaraswamy Odd Ramos-Louzada Burr XII (KumORLBXII) distribution was introduced and its parameters were determined with the maximum likelihood estimation (MLE) technique. Monte Carlo simulation was performed and the numerical results were used to evaluate the MLE technique. The proposed probability distribution's significance and applicability were empirically demonstrated using various complete and censored datasets on the survival times of cancer and diabetes patients. The analytical results showed that the KumORLBXII distribution performed well in practice in comparison to its sub-models and several other competing distributions. The new KumORL-G for diabetes and cancer survival data is found extremely efficient and offers an enhanced and novel technique for modeling survival datasets.

A new hybrid odd exponential-Φ family: Properties and applications

The paper introduces the hybrid odd exponential-Φ (HOE-Φ) family, a novel framework for generating a continuous distribution characterized by an additional parameter. The extensive statistical properties of this family are derived and explored in detail. Parameter estimation is performed using the maximum likelihood estimation technique. The efficacy and versatility of the proposed model are demonstrated through a comparative analysis involving two distinct real-world datasets.

The Mediating Role of Career Futures between Prospective Teachers’ Career Calling and Career Adaptability

This study aims to evaluate the mediating role of career futures in the relationship between prospective teachers' career adaptability and career calling. This aim was carried out according to the quantitative research patterns relational model. The sample of this study consists of 225 prospective teachers between 18 and 39 years of age ( = 21.49, SD = 1.81). Data were collected using the Career Adapt-Abilities Scale, Career Futures Inventory, and Career Calling Scale. The two-stage Structural Equation Modeling technique tested the hypothetically determined model. Accordingly, the measurement model was first examined. Afterwards, the structural model was tested. In this model test, the Maximum Likelihood estimation technique was used. Bootstrapping analysis showed that this indirect effect was significant in 1,000 bootstrapping samples. As a result of the model test, it was determined that the career futures partially mediate the relationship between career adaptability and career calling. In line with the findings of this study, recommendations for future researchers and practitioners are presented

Perceived political corruption and tax compliance intentions among private enterprises in emerging economies: The roles of tax morale and tax tolerance in Ghana

AbstractCorruption has been identified as a major debilitating factor in tax revenue generation, especially in developing countries. This study examines the effects of perceived political corruption on tax compliance intention and investigates the mediating roles of tax morale and tax tolerance among private businesses in Ghana. Data were collected with a questionnaire from 1133 private firms randomly sampled in the Greater Accra Region. Estimation was done using structural equation modelling with maximum likelihood estimation techniques. The study finds that perceived political corruption diminishes tax morale, tolerance and compliance intentions. Moreover, tax tolerance and tax morale enhance compliance intentions. Again, tax morale and tax tolerance partially mediate the relationship between perceived political corruption and tax compliance intention, reducing corruption's negative effects on intentions to comply with tax obligations. Finally, these tax compliance behaviours vary with education, age of business, sectors and firm size. The government of Ghana needs to focus on voluntary tax compliance by convincingly fighting corruption, especially by tightening fiscal loopholes.

Enhancing word of mouth in the quick service restaurants: role of perceived brand globalness and localness

Purpose Due to changing consumer thinking patterns and market dynamics, the quick service restaurant (QSR) industry has changed dramatically in the past few years. Considering this, this study aims to examine the influence of perceived brand globalness and perceived brand localness on consumer word of mouth through brand attitude by considering consumer ethnocentrism and perceived brand origin as moderators. Design/methodology/approach This study obtained 750 responses from Turkish consumers through a survey and analyzed the data using the maximum-likelihood estimation technique with structural equation modeling. Findings This study discovered that perceived brand globalness and perceived brand localness are critical components that drive brand attitude, influencing consumers' WOM toward global and local QSR brands. Similarly, perceived brand globalness and perceived brand localness are important brand attributes influencing consumer WOM. Importantly, this study found the significant effects of perceived brand origin on brand attitude mainly toward perceived local brands compared to global QSR brands. Although this study did not uncover the influence of consumer ethnocentrism as expected. However, these insights may assist global and local managers to rethink their strategies toward Turkish consumer settings. Research limitations/implications This study was conducted exclusively in Turkey. However, additional studies in other countries, such as the comparative Asian versus European consumers' perspectives, may be considered to generalize the findings. Practical implications This study provides recommendations to global and local managers to support them in designing and executing several brand positioning strategies in the QSR industry. Originality/value This novel study contributes to the accessibility diagnostic theory and signaling theory by examining consumers' perceptions of local and global brands.

Antecedents of Employee Creativity and Its Influence on Job Performance

This study aims to determine the relationship between the antecedents of employee creativity and its effect on job performance in Police personnel in Surakarta Police Station. This research method uses quantitative methods with a total population of police personnel in Surakarta police station totaling 1083 personnel. The sampling technique of this study will refer to the criteria proposed by (Joseph F Hair et al., 2014) which are based on the Maximum Likelihood Estimation (MLE) technique. A good sample number according to MLE ranges from 100-200 samples, therefore the expected sample number is a minimum of 100 respondents and a maximum of 200 respondents (Joseph F Hair et al., 2014). Specifically for this study, the sample used was around 180-200 respondents at the Surakarta City Police Station. Data collection techniques use questionnaires that will be given randomly to the object of research. Data analysis tools using Smart-PLS applications. The results showed that Mindfulness has a significant effect on the creative process of members, Support perception of the organization has a significant effect on the meaning of work, Proactive personality has a significant effect on the meaning of work, Involvement of the creative process has a significant effect on employee creativity, Meaning of Work has a significant effect on employee creativity, Mindfulness has an indirect effect on employee creativity, in the mediation relationship First perception of support The organization has an indirect effect on employee creativity through the meaning of work, Both proactive personalities have an indirect effect on employee creativity through the meaning of work, Employee creativity has a significant effect on job performance

Renyi entropy based design of heavy tailed distribution for return of financial assets

It is well-known that returns of financial assets exhibit heavy tail property and there has been no distribution which can reliably capture this characteristic so far. To contribute to the solution of this problem, we derive a new heavy tail distribution using the maximum entropy principle for Renyi entropy under the absolute moment constraints. Our newly derived distribution with two shape parameters forms a family of distributions. They are smooth, scaleable, symmetric and may be heavy tailed if their shape parameter attains the appropriate value. As a result, parameters of this distribution can be estimated by maximum likelihood estimation technique. The ability of the derived distribution to model the heavy tail property of financial assets is verified on a range of financial instruments. The results we obtained show that it can be a better option for modeling the returns of financial assets compared to other well-known heavy tailed distributions.

Exponentiated gamma constant-stress partially accelerated life tests with unified hybrid censored data: Statistical inferences

This study considers one of the partially accelerated life testing models, mainly a constant-stress approach with unified hybrid censored data. The data represents the lifetime of units following the exponentiated gamma distribution (EGD). The estimation techniques for the parameter of EGD and the acceleration factor are discussed, along with the maximum likelihood estimation technique, the Bayesian and E-Bayesian estimation. Utilizing the squared error loss (SEL) and LINEX loss functions, the E-Bayesian and Bayesian estimates are generated. For deriving the estimates of Bayesian and E-Bayesian approaches, we use the MCMC method. Real-world data is also provided for illustration. Furthermore, unified hybrid censoring schemes (HCSs) are examined to determine their effectiveness in the light of the results. In the end, we compare among the proposed methods during their findings. The study's findings demonstrate how well the suggested techniques work when utilizing unified HCS. The study also makes clear how important it is to adopt the CSPALT model, as it yields more failed things in a shorter amount of time than items with typical use situations. In conclusion, it is suggested that this research be used to tasks like industrial product reliability testing.

Do tax burdens and currency outside banks drive economic development? Empirics from Ghana

PurposeSome African policymakers have turned their attention towards electronic transaction levy (e-levy) to maximise tax revenues in recent years due to the inability to meet revenue targets. However, some argue that the implementation of an e-levy will increase the tax burden (TB) and the currency outside banks (COB). Primarily, this paper examined the effects of the TB and COB on economic development as well as the impact of institutional quality on moderating the nexus.Design/methodology/approachThis paper used structural equation modelling (SEM) and maximum likelihood (ML) estimation techniques on quarterised data from 1996 to 2020.FindingsThe results show that the TB negatively impacts gross domestic product (GDP) per capita and urbanisation but positively affects the Economic Freedom of the World Index (EFWI). The COB impacts EFWI, GDP per capita and urbanisation positively. Institutional quality moderates the TB and the COB, establishing positive relationships with the economic development indicators.Practical implicationsThe findings strongly imply that the arguments that TB and COB are catalysts for tax evasion and corruption lack substantial empirical evidence.Originality/valueThe examination of the econometric impact of the COB on economic development is one of the first studies in the field. The paper recommends that to drive economic development and accelerate sustainable development goals (SDGs) achievement, tax revenues should be channelled into the productive sectors of the Ghanaian economy.

A decreasing failure rate model with a novel approach to enhance the artificial neural network's structure for engineering and disease data analysis

The study focuses on key metrics used to examine the characteristics of a lifetime random variable distribution in reliability and survival theory research. In this analysis, metrics including the probability density function time, mean residual lifespan, mean time between failures, hazard rate, and reliability function are essential. The focus of the inquiry is these important parameters in relation to the Burr-Hatke exponential model specifically. The study focuses on key metrics used to examine the characteristics of a lifetime random variable distribution in reliability and survival theory research. In this analysis, metrics including the probability density function time, mean residual lifespan, mean time between failures, hazard rate, and reliability function are essential The focus of the inquiry is these important parameters in relation to the Burr-Hatke exponential model specifically. A key component of the research is a comparison of the outcomes from the artificial intelligence approach and those from conventional literature-based methodologies. This comparison study sheds light on how well the artificial neural network framework performs while evaluating the Burr-Hatke exponential model's technical features. The study allows a comprehensive analysis of the training and prediction capabilities of the growing neural network by calculating multiple performance measures. This comprehensive strategy improves our comprehension of the model's survival traits and reliability, offering significant contributions to the larger field of study. The network structure's mean square error was estimated to be 5.19E-04, and its coefficient of determination value was 0.99987 for the first neural network model. For the second neural network model, the coefficient of determination value was 0.99999 and the mean square error value was 4.58E-06. The outcomes amply revealed the neural network structure's extraordinarily high prediction accuracy and the degree to which the prediction outputs agree with those of the Maximum Likelihood Estimation technique.

Prediction of Solar Ultraviolet Radiations Using Random Matrix Theory

Purpose: The correlation of different samples can be described by analytical models such as random matrix theory. In this study, we tried to describe the correlation of different types of ultraviolet values in different months, weeks, and hours to get a significant relationship of special times, which one needs to get enough intensity of the sun or avoid getting sunburn.
 Materials and Methods: To this aim, we focused on the hourly and daily mean amounts of ultraviolet A, B, and C intensities of solar radiation in Tabriz urban area were measured during a full year of 2017-2018. We used such ultraviolet values which are measured at the same hour of the day to satisfy the same symmetry criteria which are necessary in random matrix theory. These data are unfolded and classified in different sequences to analyze in the nearest neighbor spacing distribution framework via the maximum likelihood estimation technique.
 Results: Strong correlation is yielded for daily values of UVA in comparison with the other types of ultraviolet radiations. Also, we considered the dependence of correlation degrees of these three types of ultraviolet to average temperature and humidity at different months.
 Conclusion: The results propose more correlation of UVA indices in August while such correlation of UVC radiations are yielded in December.

New Odd Generalized Exponentiated Exponential-G Family of Distributions

In the last decades, many researchers have developed new methods for generating families of distributions. These generators are obtained by adding one or more extra shape parameter(s) to the baseline distribution to achieve more flexibility for modelling real lifetime data sets. The additional parameter(s) has been proven useful by obtaining tail properties and improving the analysis from the goodness-of-fit for the families of distributions under study. In this paper, we proposed a new family of distributions called the New Odd Generalized Exponentiated Exponential-G Family of Distributions. The statistical properties are derived, and the maximum likelihood estimation technique is described for the proposed new family of distributions. The new models' performance is illustrated by numerical analysis using a real-life dataset, and the dataset shows that the new models offer a better fit compared to other competing models.

Effect of COVID-19 Pandemic on Labour Market Outcomes in Kenya

Purpose: This paper investigated the effect of COVID-19 pandemic on the labour market outcomes in Kenya, using quarterly labour force survey data (2019 to 2021), 2019 census data and COVID-19 wave II of a novel high-frequency phone survey collected from May 30th to June 6th, 2020. The study also estimated the impact of COVID-19 pandemic on unemployment rate by using the quarterly Labour Force Survey data from 2019 to 2022.
 Methodology: The study applied probit models because of the non-linearity of the variables. More specifically, the dependent variables are binary terms, hence maximum likelihood estimation techniques are the most appropriate.
 Findings: The findings of the study reveal that COVID-19 was associated with a 25-percentage point’s increase in the unemployment rate in the country with certain counties witnessing the highest spikes in unemployment. Although there has been some recovery, the unemployment rate remains elevated compared to pre-pandemic levels, signifying ongoing labour market challenges. COVID-19 impact on the labour market varied significantly by sector. The education sector experienced a higher incidence of employment loss, while income losses were more pronounced in sectors such as mining, manufacturing, retail, transport, and construction. Particularly, women workers in the retail and transport sectors faced twice the likelihood of income loss compared to their male counterparts. The pandemic disproportionately affected low-income workers and casual labourers, accentuating income inequalities in the labour market. Casual workers, who often lack labour protections and benefits, were significantly impacted.
 Unique Contribution to Theory, Practice and Policy: Considering these findings, the study provides several policy recommendations: Implementing targeted interventions and support programs in counties with the highest unemployment spikes. These interventions can include skill development programs to enhance workers' employability and tailored employment and income support strategies adapted to each county's unique economic circumstances. Prioritize the adoption of Information and Communication Technology (ICT) to enhance resilience and adaptability across sectors. Governments, both at the national and county levels, should prioritize policies aimed at enhancing the incomes of disadvantaged households. Through expanding access to social protection programs, enforcing minimum wage regulations, and improving working conditions, especially for casual workers and low-income households. The study's insights provide valuable guidance for shaping future economic recovery strategies, particularly in the event of economic shocks in the labour market.

Discretionary Accrual Earnings Management in a Sub-Saharan African Economy

This paper investigates the existence of discretionary total accrual earnings management and working capital accrual earnings management in Nigeria which is Sub-Saharan Africa's largest economy. Unlike previous studies that use the Ordinary Least Squares (OLS) technique that does not control for endogeneity that can invalidate results, our empirical methodology implements the maximum likelihood estimation that controls for endogeneity to estimate economically plausible levels of discretionary accrual earnings management. Our results find that discretionary accrual earnings management in Nigeria is characterized by statistically significant, tax-induced income-reducing discretionary accruals. This study contributes to the literature as it is the first emerging market, African and Sub-Saharan African economy study to the best of the researchers’ knowledge that estimates discretionary accrual earnings management for listed non-financial firms using various models based on the maximum likelihood estimation technique to control for endogeneity. The findings inform the political economy of economic development in Sub-Saharan Africa, particularly as it relates to how firms, especially multinationals, are motivated to extract rents from African economies. The results have policy implications for accounting regulatory interventions by policy and standard setting bodies that mitigate and/or eliminate capital flight, tax evasion, aggressive tax avoidance and financial corruption practices in Sub-Saharan African economies.

Application of Principal Component Analysis and Maximum Likelihood Estimation Method to Identify the Determinant Factors Intention to Use of Paylater in E-Commerce

This research aims to analyze which factors are most determining in influencing interest in using pay-later. This type of research is multivariate analytical research using a quantitative approach. All pay-later service customers who engage in e-commerce make up the study's population. Purposive sampling and non-probability sampling were employed in the sample-taking process based on this study. There are one hundred samples. Researchers have made observations to gather data. The author intends to employ principal component analysis and maximum likelihood estimation techniques for data analysis. Following the completion of the study, the following findings were made. The principal component analysis method's RMSE value is 0.02. The maximum likelihood estimates method's root mean square error is 0.015. Therefore, it can be said that the principal component analysis approach is appropriate for usage because it can explain factor formation due to its strong factor loading value and low RMSE value. The influence of advantages is the most important element determining interest in pay-later utilizing the principal component analysis method since it has the highest loading factor value in decision-making.

Zero-Inflated Poisson Regression Modeling of Plant Protein Consumption

Summary This research fitted a discrete distribution for modeling count data. Specifically, Zero-Inflated Poisson (ZIP) regression was used to model plant protein consumption by 400 randomly sampled individuals in Wukari. The data was collected by questionnaire. The ZIP regression model was used based on its ability to model data with excess zeros present in the collected data. Variables considered and used for the analysis are Age, Body Mass Index, Blood Pressure, Occupation, Gender, Weight, Height, Body Reaction, and Consumption Class. The parameters of the ZIP model were estimated using the maximum likelihood estimation technique. The model was tested for Goodness of Fit (GoF) using deviance, scaled deviance, Pearson–χ 2 and scaled Pearson–χ 2 statistics. The results obtained showed that Age, Gender, and Reaction were significant at 5%, and the GoF tests revealed that the Zero-Inflated Poisson regression produces a good fit and is a good model for overcoming the overdispersion effect.

Non-Mixture Cure Model Estimation in Bladder Cancer Patients: A Novel Approach with Exponentiated Weibull Exponential Distribution.

Cure models are frequently used in survival analysis to account for a cured fraction in the data. When there is a cure rate present, researchers often prefer cure models over parametric models to analyse the survival data. These models enable the ability to define the probability distribution of survival durations for patients who are at risk. Various distributions can be considered for the survival times, such as Exponentiated Weibull Exponential (EWE), Exponential Exponential (EE), Weibull and lognormal distribution. The objective of this research is to choose the most appropriate distribution that accurately represents the survival times of patients who have not been cured. This will be accomplished by comparing various non-mixture cure models that are based on the EWE distribution with its sub-distributions, and distributions distinct from those belonging to the EWE distribution family. A sample of 85 patients diagnosed with superficial bladder tumours was selected to be used in fitting the non-mixture cure model. In order to estimate the parameters of the suggested model, which takes into account the presence of a cure rate, censored data, and covariates, we utilized the maximum likelihood estimation technique using R software version 3.5.7. Upon conducting a comparison of various parametric models fitted to the data, both with and without considering the cure fraction and without incorporating any predictors, the EE distribution yields the lowest AIC, BIC, and HQIC values among all the distributions considered in this study, (1191.921/1198.502, 1201.692/1203.387, 1195.851/1200.467). Furthermore, when considering a non-mixture cure model utilizing the EE distribution along with covariates, an estimated ratio was obtained between the probabilities of being cured for placebo and thiotepa groups (and its 95% confidence intervals) were 0.76130 (0.13914, 6.81863). The findings of this study indicate that EE distribution is the optimal selection for determining the duration of survival in individuals diagnosed with bladder cancer.

The Type II Topp-Leone Inverse Power Lomax distribution with Simulation and Applications

In this work, we present a four-parameter lifetime model that can be used to model reliability issues, fatigue life studies, and survival data called the Type II Topp-Leone Inverse Power Lomax distribution. It has the Type II inverse Lomax, Inverse Power Lomax, and Inverse Lomax distributions as sub-models. Some of its statistical properties, including complete and incomplete moments, generating functions, characteristics functions, mean residual life, mean inactivity time, Renyi entropy, Tsallis entropy, order statistics, stress-strength reliability, and weighted probability moment, have formal formulas that we have developed. The model's parameters are estimated using the maximum likelihood estimation technique. The effectiveness of maximum likelihood estimators is evaluated in terms of absolute bias and simulation study standard error. Two lifetime data sets are used to demonstrate how the new model can be applied. Using the same comparative criteria, the proposed distribution offers a better fit than a few well-known distributions.