Department of Physical Science

Inference-Driven Logistic Regression Approach for Identifying Risk Factors of Myopia

Thu, 01 Jan 2026 00:00:00 GMT

Inference-Driven Logistic Regression Approach for Identifying Risk Factors of Myopia Kayathiri, T.; Kayanan, M.; Wijekoon, P. Myopia is a prevalent refractive error among children and adolescents and represents an increasing global public health issue. The present study aims to identify key risk and protective factors associated with myopia onset through statistical inference. A real-world myopia dataset from the R statistical package, which has been referenced in prior research for theoretical validation and factor identification, was analyzed. This investigation extends previous work by utilizing statistical inference to distinguish between variables that elevate myopia risk and those that confer protection. The dataset comprises 618 individuals who were non-myopic at baseline and were observed over a period of at least five years. During this period, 17 clinical and behavioral variables were recorded. Predictor variables include age, spherical equivalent refraction (SPHEQ), axial length (AL), anterior chamber depth (ACD), lens thickness (LT), vitreous chamber depth (VCD), and time spent on activities such as sports (SPORTHR), reading (READHR), computer use (COMPHR), studying (STUDYHR), and watching television (TVHR). The binary response variable was coded as 1 for myopic and 0 otherwise. Logistic regression coefficients were estimated using maximum likelihood estimation (MLE). Findings indicated that age, SPHEQ, AL, SPORTHR, STUDYHR, and TVHR exhibited negative coefficients, suggesting that increases in these variables are associated with a reduced risk of developing myopia. Conversely, positive coefficients for ACD, LT, VCD, READHR, and COMPHR point to elevated myopia risk. At the 5% significance level, SPHEQ and SPORTHR emerged as statistically significant predictors, while READHR and STUDYHR achieved significance at the 10% level. At the 90% confidence interval for odds ratios, SPHEQ, SPORTHR, and STUDYHR show protective effects (odds ratios < 1), whereas READHR is linked to greater risk; these protective effects remain significant at the 95% level. In summary, reduced reading time, increased participation in sports and studying, along with specific ocular measurements, may mitigate the risk of myopia development.

On the development of the adjusted multinomial logistic Liu estimator

Thu, 01 Jan 2026 00:00:00 GMT

On the development of the adjusted multinomial logistic Liu estimator Kayathiri, T.; Kayanan, M.; Wijekoon, P. The multinomial logistic regression (MNLR) model is a widely used statistical tool for predicting categorical response variables with more than two outcomes, based on multiple predictor variables. It finds applications across various fields, including healthcare, social sciences, and marketing. The maximum likelihood estimator (MLE) is the standard method for estimating parameters in MNLR models. However, when predictor variables exhibit multicollinearity, the MLE becomes inefficient, resulting in inflated variances and unstable coefficient estimates. To address this issue, a limited number of biased estimators have been proposed in the literature. This study aimed to develop an efficient estimator that reduces the impact of multicollinearity in MNLR models by introducing the adjusted multinomial logistic Liu estimator (AMLLE), which improves estimation accuracy and stability. A Monte Carlo simulation study was conducted to evaluate the performance of the MLE, multinomial logistic ridge estimator (MLRE), multinomial logistic Liu estimator (MLLE), almost unbiased multinomial logistic Liu estimator (AUMLLE), and AMLLE under moderate to high multicollinearity. The simulations considered a wide range of sample sizes and varying levels of the response variable. The findings indicated that the proposed estimator, AMLLE, outperformed the existing estimators in all scenarios considered. The relative efficiency of AMLLE compared to MLE, based on SMSE, showed substantial improvement across different correlation values. For correlations of 0.5, 0.7, 0.9, and 0.99, AMLLE achieves efficiencies of 38.25, 46.62, 68.73, and 96.79% for n=50; 21.56, 27.03, 46.25, and 90.80% for n=100; and 2.85, 3.80, 9.04, and 41.21% for n=1000, with three response levels. The corresponding efficiencies with five response levels are 43.47, 52.16, 61.86, and 97.95%; 27.66, 33.39, 52.84, and 93.95%; and 3.76, 5.23, 12.41, and 47.05%, respectively. Moreover, increasing the sample size further enhanced the performance of the proposed estimator, while higher correlation and additional response levels tend to reduce its effectiveness. In conclusion, the adjusted multinomial logistic Liu estimator provides a reliable and computationally efficient alternative for parameter estimation in multinomial logistic regression models affected by multicollinearity. It shows strong potential for practical applications, and future research could explore its extension to high-dimensional predictor settings.

Behaviour of Rainfall Patterns in the Trincomalee District: A Time Series Modelling Approach

Sun, 01 Jan 2012 00:00:00 GMT

Behaviour of Rainfall Patterns in the Trincomalee District: A Time Series Modelling Approach Yogarajah, B.; Elankumaran, C.; Vigneswaran, R. With the advent of rapid developmental activities in the Eastern province of Sri Lanka in the post-war scenario, the government’s mandates focus on reviving plans for the agricultural sector to meet the growing demands of the nation. Understanding the behavior of the climatic parameters of a geographical area is a prerequisite for any effort towards developing the agriculture sector. Climatic variability, especially the unpredictability of rainfall regimes is a major constraint for agricultural planners when it comes to deciding the time of planting in the Trincomalee district. The aim of this paper is to explain and analyze the temporal behavior of long-term monthly retrospective rainfall data of the district using ARIMA technique. This paper focuses on a time-series modeling approach to understand the behavior of rainfall patterns for the period from January 1952 to December 2009. The ARIMA model analysis proved to be a very valuable technique in forecasting climatic trends for Agro-environmental planning (Sabita Madhvi Singh, 2012). Rainfall time series data are analyzed using ARIMA statistical techniques to study the annual and seasonal trend of climates, fluctuation and variability. Various seasonal ARIMA models were tried in this respect. Key findings indicate that the rainfall patterns in the study area modeled as ARIMA(1,0,0)(1,1,1)12, as such the rainfall predominantly depending on nonlinear trend and seasonal pattern of order 12 with the autoregressive of order one combined with lag12 process. This indicates that comprehensive forecasting model for rainfall in Trincomalee district is arrived. Further research is needed to focus on the influences of non-endemic and regional-to-global climatic phenomena.

Occupational health and safety in the noisy work environments; from the perspective of lathe, saw and grinding mills in vavuniya district

Wed, 01 Jan 2025 00:00:00 GMT

Occupational health and safety in the noisy work environments; from the perspective of lathe, saw and grinding mills in vavuniya district Kuhanesan, S.; Arjunan, K. Occupational health and safety (OHS) are the key aspects of human concern. OHS aims for highest degree of physical, mental and social well-being of workers in all occupations. Recent technical advancements have brought up sophisticated machinery and equipment to both industrial production sector and to services and commerce. As such, mechanization results in greater vulnerability on the exposure to higher noise levels. Auditory and non- auditory health effects caused by the occupational noise exposure has broadly reported so far. However, in Sri Lanka, such studies on occupational noise exposure and subsequent health effects have not been extensively studied yet. In this study, we aimed to assess the noise levels in the working environment including grinding mills, Lathe/ metal processing and saw mills in the Vavuniya district and the views/ perception of workers on the impacts of occupational exposure to higher noise levels. Noise level measurements were obtained from randomly selected eight grinding mills, four lathe shops and 4 saw mills in Vavuniya district. TENMA 72-947 model sound level meter with the measurement range of 30dB to 130dB and the datalogging facility was used to take noise level measurements. Semi- structured interviews were conducted at respective measurement places to gather information about the status/views/ perception of workers about the auditory and non- auditory impacts of exposure to higher noise levels. Measured noise levels during the operation of machineries in the Grinding mills, Lathe and Saw mills were in the range of 90 to 98 dB, 78 to 94 dB and 92 to 100 dB respectively. Measured noise levels were well-above the World Health Organization (WHO) recommended noise level for industrial area (70 dB). Workers were well aware about the impacts; however, they were reluctant to have personal protective equipment (PPE) by stating the reasons including duration of exposure, cost for noise control setups and PPE and convenience. Thus, appropriate strategies to be implemented to avoid occupational health and safety related complications of workers who experience exposure to higher noise levels.