Browsing by Author "Liyanage, U. P."

Now showing 1 - 17 of 17

Algorithm to identify the original web links and suggest optimized mirror links for download content within a web page.
(International Research Symposium on Pure and Applied Sciences, 2017 Faculty of Science, University of Kelaniya, Sri Lanka., 2017) Karunarathna, G. L. S. I.; Liyanage, U. P.
Invention of the Internet has become a revolutionary change to the world. People use different technologies to connect to the Internet. Surfing Internet has become a stressful activity due to the existence of various spams and redirections. Consequently, internet surfers suffer from wasting time and money on in relevant web contents every day. Further, looping redirections caused to distract many internet surfers all over the world. Though the adware blockers come to the stage for preventing unwanted ads, it does not come with handy solution for assisting web surfers to direct the desired web content or resource. At the same time, there can be lots of mirror links, which are available for refer the same web content or resource. If the web surfer is provided desired content targeted and optimized mirror link/s that has minimum traffic and higher bandwidth with minimum estimate time to download the file, it will be much more useful. The purpose of the research is to achieve the solution for suggest original link to download and provide optimized download link. A chrome extension, which is run in chrome browser, is built with all the proposing components and algorithms in order to proof of the concept. Through this highlight original resource link in the web page and pointing fake/redirect links in the web page. Define an algorithm to suggest optimized mirror link to download among the original mirror links. The tool supposed to cache all the metadata of the referred links and validate links time to time with update latest state of the links. The ultimate objective is to derive an algorithm to avoid fake web redirection links and download resources in cost effective manner. Additionally, the software solution implementing this algorithm protect the computer system by avoiding the links that contain harmful malwares and virus. This proposed software solution will develop as platform independent chrome extension and deploy to ensure the optimum and safe internet surfing.
Anthurium disease detector using machine learning techniques
(Faculty of Science, University of Kelaniya, Sri Lanka, 2021) Ranasinghe, K. M. N. S.; Liyanage, U. P.
There are many problems with planting Anthurium due to diseases. The lack of knowledge about the diseases and identification of the disease correctly, are the major challenges that face by the farmers. Further, some of diseases have no cure, yet must be destroyed before they can be spread. The ways and means of obtaining the essential assistance, for example may be from responsible parties such as Department of Floriculture or Agriculture in Sri Lanka, are limited due to the fact that there is no flexible mechanism to approach their resource personals in efficient manner. Thus, treatment of ill plans becomes hard problem. To overcome these practical difficulties faced by farmers, this study focus to build a mobile application using Android Studio Software to detect the disease through image processing and computer aided models, and thereby, allows farmers to apply disease treatment steps as soon as possible. Using the mobile application, farmers allows to take pictures of disease plant parts, and send for disease detection programs. These programs analyses the images using appropriate machine learning techniques and gives a feedback concerning the disease the plant has. Indeed, correct identification of the plant disease leads to early treatment, and hence the better curing possibilities. The system allows to determine diseases using the images of the Anthurium leaves, flower body, flower nose and the roots of the plants. The system allows to recognize three diseases namely, Bacteria Blight, Rhizoctonia Root Rot and Black Nose Disease. To determine disease correctly, images from all the above mention plant parts are needed. As the disease detection techniques, algorithms using the Convolutional Neural Network (CNN) have been utilized. In particularly, a sequential CNN model namely, LeNet, have been trained and tested using 2962 images. Further, all the images are transformed to gray-scale images to improve the classification rate of CNN algorithms. The models are trained with 10,15,20,30,40,50,75,80 and 100 epochs, and the scenario with 80 epochs performed best in terms of both accuracy and loss values and had the best curve. Based on the image set that has been used, Bacteria Blight, Rhizoctonia Root Rot and Black Nose Disease have been detected with the accuracy of 96.5%, 99% and 98%, respectively. The predicted time is on average less than one second using average computer power utilized in the back end of the system. This accuracy is sufficient for successfully detecting the plant diseases, and thereby, the system that has been engineered will be beneficial for the farmers to manage healthy plant nursery.
Application of Life Cycle Perspective Costing (LCC) Use in European Union (EU) as a Green Procurement Tool for Cost-effective Public Procurement in the Health sector, Sri Lanka: A Survey-based Study
(Faculty of Commerce and Management Studies, University of Kelaniya Sri Lanka, 2023) Dinesh, K. L. S.; Liyanage, U. P.; Wijeyaratne, W. M. D. N.
In Sri Lanka public procurement guidelines should be followed by all government agencies, when they are required to procure goods, works, and services. The Objective of such guidelines is to ensure the value for money of the process. The Ministry of Environment, Sri Lanka has issued a national policy on sustainable consumption and production for Sri Lanka. Two sections of this policy document are reserved for public procurement and Health sectors. According to the policy document, the Ministry of Health should ensure sustainable practices at all levels in the health sector to be transformed into green work set up by 2030. The policy suggests applying sustainable public procurement (SPP) practices in all sectors and for each product or service that has a significant cost-saving impact. This paper mainly discusses Life Cycle Perspective Cost (LCC) and other Green and Sustainable evaluation practices regulated in health sector institutes in the European Union (EU) and reviews the possible adoption of those green and sustainable concepts to public procurement processes in Sri Lanka for cost-effective Public Procurement. The Adoption of green would maximize the value of public money while minimizing damage to the environment and maximizing economic, and social benefits to the public health sector in Sri Lanka.
Assessment of awareness and practice on plastic water bottle usage among undergraduates of University of Kelaniya
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Dewrandi, M. B. K.; Gayathri, T. K.; Lakshika, S. M. D.; Sriskantharajah, S.; Tishopana, S.; Liyanage, U. P.; Rajapaksha, G.
Bisphenol A (BPA) is highly used in the production of polycarbonate plastics. BPA migrates from polycarbonate containers, leading to human BPA exposure. BPA is an endocrine disruptor, which perturbs natural endocrine pathways. BPA leaching from plastic products is affected by pH, high temperature, product quality, washing practices and age of the product. Even though plastics are popular in the Sri Lankan market, scientific research on their usage pattern and public awareness on health risks are poorly known. Therefore, this study was aimed at investigating plastic water bottle usage, consumer practices, and public awareness on BPA using a questionnaire survey. A questionnaire was developed and validated to obtain information on types of plastic water bottles used, duration of use, price of the product, cleaning practices, nature of beverages stored, knowledge on plastic associated health risks including BPA and knowledge on safe use of plastic bottles. The online questionnaire was administered among 502 undergraduates in the University of Kelaniya during the period of September–November 2023. A majority of 348 students (69.3%) use plastic water bottles and nearly 38.5% of undergraduates were at risk of BPA exposure. Majority of the students were at increased risk of exposure to BPA and other plastic-related chemicals due to unfavorable practices such as storing at high temperature (54.9%), storing acidic beverages (40.3%) and practicing harsh washing conditions (46%). A total of 289 students (83.0%) were aware on leaching of toxic chemicals from plastic bottles. Despite, majority (70.4%) lacked awareness on major plastic related chemical, BPA and its health implications. Out of the plastic users, only 28.2% were aware on both plastic-associated chemicals and on BPA migration. 191 undergraduates (54.9%) were aware of toxic chemicals associated with plastics but were not aware of BPA. Also 15.5% of users were not aware on any type of risk associated with plastic usage. Although polycarbonate and non-recycling number bottle users are of increased risk of BPA exposure, out of polycarbonate plastic users, 70.8% were not aware on BPA. Also, out of the users of plastics without any number, 70.9% were not aware on BPA. As per the survey, plastic water bottles were popular among undergraduates in the University of Kelaniya. However, awareness on plastic-associated health risks and consumer practices were unsatisfactory.
A clustering based quantitative approach on selecting companies in an investment portfolio in Colombo Stock Exchange
(4th International Research Symposium on Pure and Applied Sciences, Faculty of Science, University of Kelaniya, Sri Lanka, 2019) Kumara, M. S. M. S.; Liyanage, U. P.
Portfolio management is a powerful concept in financial sector, heavily studied by both investors and researchers. Conventionally, investment portfolios on stocks available in stock markets, constitute by set of stocks belonging to numerous companies and their associated allocations. Additionally, the standard portfolio procedure results the optimum allocation of shares of selected set of companies, with the minimum risk. Nevertheless, the selection of companies in a portfolio is utterly depends on the experiences as well as the gut-feelings of investor or the broker. Thereby, this selection criterion is essentially conditional on qualitative measures that have no numerical justifications. This research aims to introduce a quantitative approach towards selecting companies into a portfolio based on their historical data so that the portfolio optimization procedure can overcome the qualitative bias. The analysis has been conducted using the stocks belonging to companies registered at the Colombo Stock Exchange (CSE), Sri Lanka. The data consisting of daily share prices of 291 companies registered at CSE for the period 2012-2016. The company risk is measured by the volatility of its stock prices over the time. In standard portfolios, there is a mix of companies with various risks. Technically, here a novel mechanism to determine composition of companies in such portfolio based on risk levels has been introduced. Different risk levels are determined by using K-Mean clustering technique applied on the volatility of companies. Since the history of stock prices essentially determine the risk levels, the volatility has been captured so that it would reflect the historical behavior of the company’s stock prices. Consequently, volatility has taken as a vector that has elements consisting of corresponding variance measured by quarterly basis. Number of quarters resulting the dimension of volatility-vector, is selected as four in this study. The clustering procedure determining the risk levels is based on the volatility-vectors computed on each company, used to obtain five classes of companies with different risk levels. Sorting the classes by mean risk from low to high, allows to select the composition of companies in the considered portfolio. In this research, to establish the portfolio, proportion of companies (0:3:4:3:0) belonging to classes from low to high risks, are selected. This selection allows to balance the risk among companies within the portfolio. The study shows that portfolios have higher return can be constructed by such selections from the clusters appropriately. Further investigation of selection criterion based on such proportions have been analyzed
Distribution of heavy metals across different tissue types in Etroplus suratensis from Mahakanadarawa reservoir: Investigating dietary implications for CKDu in Sri Lanka
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Perera, P. L. R. A.; Perera, W. P. R. T.; Liyanage, U. P.; Premaratne, W. A. P. J.; Liyanage, J. A.
The escalating cases of chronic kidney disease of uncertain etiology (CKDu) in rural communities, especially in the Maradankulama-Mahakandrawa region of Sri Lanka, stipulate a critical public health investigation. Given the significant dietary reliance on Etroplus suratensis due to geographical and logistical complexities in obtaining sea fish, bioaccumulated toxic metals in this species can be a potential risk factor for CKDu. This study investigated the heavy metal content (As, Pb, Cd, Cr, Ni) in the gut, gill, and muscle tissues of E. suratensis sourced from the Mahakanadarawa Reservoir in the CKDu endemic Mihintale region. To obtain representative E. suratensis samples from across the Mahakanadarawa Reservoir, fish were collected using a randomized sampling method. Multiple random points within the reservoir were pre-determined and fish were caught at each location by local fishermen using appropriate techniques. The captured fish were identified as E. suratensis and 36 samples were separated as gut, gills, and muscles for the toxic metal analysis. The heavy metal content (As, Pb, Cd, Cr, Ni) in these samples was then analysed using inductively coupled plasma mass spectrometry (ICP-MS). To thoroughly compare metal concentrations across tissue types, the non-parametric Kruskal-Wallis ANOVA test was utilized followed by post-hoc Tukey HSD tests. The non-parametric ANOVA assessed statistically significant differences in metal levels between muscle, gill, and gut tissues. Post-hoc Tukey tests then enabled pairwise comparisons between each tissue type to determine which specific metal concentrations differed. There were significant differences between tissue types for concentrations of As, Cd, Cr and Ni (p<0.001). Post-hoc Tukey tests showed As, Cr, Cd and Ni were significantly higher in gut compared to that of muscle and gill tissues (p<0.05), with the order Gut>Gill>Muscle. Pb did not deviate significantly across tissues, however, mean Pb concentrations exceeded the WHO/FAO permissible limits for dietary intake in all tissues, while Cd remained within acceptable levels. Compared to gut and gills, muscle contained relatively low concentrations of As (0.01±0.01) mg/kg, Cr (0.22±0.03) mg/kg, Pb (0.508±0.36) mg/kg, and Cd (0.015±0.13) mg/kg and Ni (0.06±0.01) Although edible muscle in some samples met regulatory limits, frequent consumption of E. suratensis from this CKDu endemic area may pose a health risk, warranting further study on geographical and seasonal variation. Ultimately, this study contributes to the growing body of evidence suggesting that bioaccumulation of toxic metals in fish poses significant CKDu risk factors.
Economic impact of COVID-19 on the total revenue of the textile and apparel export industry in Sri Lanka
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Kaushalya, H. A. T.; Priyadarashana, D. A. D. S.; Liyanage, U. P.; Hewaarachchi, A. P.; Jayamanna, J. M. A. D. A. P.; Virani, D. G. D.; Dilshan, H. R.; Viduranga, A. L.
Textile and apparel exports play a vital role in economic development in Sri Lanka. It’s approximately 44% of total exports in the country by 2024. The apparel sector has become complicated in recent times because of economic instability followed by COVID-19. The purpose of this study is to quantify the revenue impact caused by COVID-19 on the textile and apparel export industry in order to address this significant issue. This will facilitate plans by textile and apparel exporters more appropriately. The COVID-19 period has resulted a negative impact on the Sri Lankan textile and apparel industries. If the industry faces such an unfortunate epidemic in the future, it is beneficial to know the impact of COVID- 19 on the industry. Thus, the decision-makers and experts in the industry will be able to make sound decisions and plan effectively to reduce the impact. This study has considered monthly export revenues from January 2009 to April 2024. Data were collected from the Joint Apparel Association Forum Sri Lanka (JAAFSL). The revenue data before the COVID-19 period, from January 2009 to December 2019, was considered in the model-building process. First, the preliminary transformation, including the log transformation, first differencing, and seasonal differencing, were applied to obtain a stationary data set. Then, fifteen candidate models were selected based on the Akaike Information Criteria (AIC). ARIMA (0,1,1) (0,1,1)12 model has been selected, which has the lowest AIC value and MAPE value of 1.29%. Further, the model diagnosis was checked using a residual analysis. The p-value of the LjungBox Q-test is 0.148, which confirmed that the residuals are white noise. After that, the monthly revenue was predicted for the COVID-19 period using the fitted model. To quantify the impact of COVID-19 two curves were fitted to the actual revenue and the predicted revenue after COVID-19. The difference between areas under both curves was then computed. The impact has been estimated as a percentage of this difference to the predicted revenue. The results of this study arrived at the high impact of an 11.92% decrease in export revenue due to the COVID-19 pandemic on the textile and apparel export industry, with the estimated loss equivalent to 1899.1291 million USD. This figure underscores the significant impact on the export industry and the country's economy. Such understanding will assist experts and decision-makers in evolving their strategies to move out of such situations in the future, providing fresh mechanisms to avoid this kind of impact.
Estimating COVID-19 prevalence in Sri Lanka
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Erandi, J. D. T.; Liyanage, U. P.; Gunawardana, A.
Throughout the ages, man has had to face numerous crises and diseases. Among them, the COVID-19 virus can be considered as one of the most fatal diseases ever, and it has caused significant damage to the entire world. Moreover, due to the nature of the virus transfer modes, controlling the COVID-19 infection among people is a challenging task, and thereby, the spread of the virus still persists globally with less severity. Hence, an effective and accurate controlling measure is essential. The profile of the coronavirus progression in a sub-region can be changed due to numerous factors such as population density, public mobility, and available health facilities. Thus, at a time, diverse prevalence status of virus spread on different sub-regions is highly probable. This study attempts to construct a suitable sampling design to capture the prevalence of COVID-19 by modifying the stratified sampling technique to estimate the sample size adapting to the changing population of infected cases. This adaptation is essential as the increase of infected cases boosts the virus spread, and the standard sampling techniques do not address such dynamic population conditions in determining the sample size. Further, the study bridges the gap between the reported and actual infections per day, thereby giving accurate estimates of virus distribution and prevalence. The coronavirus progression over a region has a skewed pattern, and it should also be considered in the weight allocation method. Thus, the weights are determined based on the first derivative of reported infected cases. This derivative information is based on the recent dynamics of the infected cases. Consequently, larger weights were assigned when the virus progression increased, and smaller weights were assigned when the virus progression decreased. After that, the sample size for each sub-region was calculated by the modified stratified sampling method. To illustrate the accuracy of the sampling design, simulated data from different epidemic scenarios, such as community spread, cluster spread, and border spread, was used. This simulation allowed us to test the robustness of the techniques for the different states of the virus progression based on the infected cases. The sample size obtained through this dynamic sampling technique exhibits a direct correlation with the fluctuations in the number of infected cases, increasing as the infection cases rise and decreasing as they decline. In conclusion, the study results in a novel sampling technique that is sensitive to the dynamic nature of population sizes, and it can be straightforwardly applied to real-world data as well. Thus, this modified stratified sampling technique can be considered as an accurate sampling technique to capture the actual prevalence of COVID-19.
Factors influencing milk powder preferences: a comparative study of local and imported brands in Gampaha district
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Maleesha, M. A. N.; Prabhashwara, M.; Wickramage, R.; Abeygoonewardena, V.; Rathnayake, N.; Madhushani, T.; Liyanage, U. P.; Dissanayake, D. M. P. V.
Milk powder is a common choice in Sri Lanka because it's convenient, easy to store, and lasts longer than fresh milk. It is made by removing the moisture from milk, which makes it practical for storage and transport, especially in households where fresh milk isn’t always accessible. However, due to the limited supply of locally produced milk powder, many people rely on imported brands. This study looks at why people in the Gampaha district, the most densely populated district in Sri Lanka, choose between local and imported milk powder. Gampaha has both urban and rural areas, making it a good place to study consumer habits. The research involved a survey given to people in three Divisional Secretary’s Divisions (DSDs) out of 13 in the district. These areas were selected using cluster sampling, and convenience sampling was used to gather the data. The sample size was set at 303, based on a pilot survey, to ensure the findings would be reliable. The results showed that price, quality, and taste were the most important factors influencing people’s choices when it came to milk powder. We used chisquare and G-square tests to confirm that these factors had a significant effect. Interestingly, household income and brand preference have not seemed to play much of a role in people’s decisions. Some respondents also pointed out that herbal drinks like Ranawara, Belimal, and Kolakenda were viewed as alternative beverages to milk powder, especially when milk was not affordable or available. While many people believe that locally produced milk powder is of better quality, they often end up choosing imported brands because they are cheaper and more widely available in stores. There’s also a strong sense among consumers that they want to support local products, but the higher cost of local milk powder makes it hard for many to buy. The study suggests that making local milk powder more affordable and available in more places could reduce the dependency on imports, benefiting not just the consumers but also the national economy by supporting local dairy production. If local milk powder were priced more competitively, it would encourage more people to choose it over imported brands. In conclusion, improving both the affordability and accessibility of local milk powder could help boost its consumption, strengthen the local dairy industry, and reduce reliance on imports, bringing economic benefits to Sri Lanka in the long run.
Forecasting phenological model for tropical forest species: Monoon coffeoides
(Faculty of Science, University of Kelaniya, Sri Lanka, 2021) Wellassa, P. W. G. S. R.; Ratnayake, R. M. C. S.; Liyanage, U. P.
Monoon coffeoides is a forest plant growing a tropical intermediate forest and pollinated by a native species of weevils in Sri Lanka. The longtime survival of both M. coffeoides and weevils is governed by its flowering and fruiting phenology. Due to a lack of phenological information implementation of proper conservation and management measures are not possible. The main purpose of this study is to analyze phenological parameters, namely, flower buds, open flowers, leaf flushing, leaf dropping, immature fruits and mature fruits of the Monoon coffeoides, to identify their correlation and variation patterns. Further, forecasting of these parameters are important in future forest management. Hence, the study has been extended to investigate the ability to forecast the parameters. Unsupervised learning techniques such as K-means clustering under Data mining are applied to identify similar behaviors among 50 trees of Monoon coffeoides. Silhouette width test was used to validate the cluster accuracy. Further, the Cross-correlation analysis was used to identify the relationships between series of phenological parameters with following delay periods as lag phases. The analysis resulted, the delay between flower buds and open flowers is 2 weeks, the delay between leaf flushing and mature fruits is 17 weeks, the delay between open flowers and immature fruits and between immature fruits and mature fruits is 5 weeks. Additionally, it was identified that the flowering and fruiting periods are varied from January to May and from March to August respectively. As the next step, phenological parameters and climate factors have been forecasted using univariate time series models. The accuracy was tested using standard tests: R-squared, Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Absolute error (MAE). The best-fitted models for each of these parameters are; flower buds: ARMA(2,3)-GARCH(1,1), open flowers : ARMA(2,3)- GARCH(1,1), mature fruits : ARIMA(1,1,1)-GARCH(1,1), immature fruits : ARMA(1,4)- GARCH(1,1), leaf dropping : ARMA(1,1), leaf flushing : ARMA(1,1)-GARCH(1,1), average temperature : ARIMA(1,2,1) and rainfall : ARMA(3,1) respectively. All the models were significant to forecast the values and thereby, these models can be used to forecast phenological parameters.
Proactive Dengue Management System Synergize by an Exponential Smoothing Model
(Research & Development Centre for Mathematical Modelling,Department of Mathematics, University of Colombo,Sri Lanka, Department of Statistics & Computer Science,Faculty of Science, University of Kelaniya, Kelaniya, Sri Lanka., 2021) Wetthasinghe, W. A. U. K.; Attanayake, A. M. C. H.; Liyanage, U. P.; Perera, S. S. N.
In a critical area like health sector centralized computer system helps to improve the efficiency of the health system. In particular, controlling an epidemic is usually difficult in developing countries. In this study we introduce a multi-platform, centralized proactive management system to manage dengue controlling activities in Sri Lanka. The system make common platform (ProDMS) for all sectors who contribute their services for mitigating dengue [1]. We mainly focused to the special feature of the system which enhance the centralized property. Cross platform environment was developed under this feature as a bridge to connect researches and general public. ProDMS is a internet base web application and researches can plug their dengue forecasting models to the system and publish their outputs as graphs through the web system. The ProDMS web application, which consisting of plug and play system architecture concepts, fully support for any statistical or mathematical model to publish its results online. In this work we use one of the univariate time series modelling approaches; namely exponential smoothing to plug with the system. This research helps to enhance efficiency of Dengue controlling process and support to generalize centralization.
A quantitative analysis of fishery industry in modelling of production, trade dynamics, and COVID-19 impact estimation
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Sandamini, R. P. K.; Tharaka, K. D. R.; Wickramasinghe, W. A. P. K. V.; Mendis, T. A. B.; Wijesekara, J. M. C. D.; Lakruwan, J. M. D. C.; Dasanayaka, O. R. G. K.; Liyanage, U. P.; Hewaarachchi, A. P.
Sri Lanka, being an island, is granted an immense maritime boundary. The fishery industry is a prominent and significant part of the Sri Lankan economy, contributing around 1.5% of the Gross Domestic Product. The objectives of this research are to identify the factors affecting fish production, analyse the Net Trade Balance (cost difference between import and export of fishery products) in the fishery industry, and estimate the impact of COVID-19 on fish production. The monthly data from 2015 to 2022 is taken from the Statistics unit of the Ministry of Fisheries, Sri Lanka. This data is analysed, and the findings revealed that fish production in the country had decreased significantly over time, and thereby, it has impacted the import quantities as well. Further, almost 80% of the total fish production of the country is contributed by marine fish production and 20% by inland fish production. The variations in fish production are largely caused by the inland fish harvest; that is marine fish production has monotone dynamics. Concerning the seasonal pattern, less production is visible during May and June due to the monsoon. However, a clear upward trend in the inland fish harvest is visible during this time due to the renewal of freshwater. According to a recent survey in the Indian Ocean, the decrement in the fish population and habitats contributed to the lower harvest of fish production. Further, it has been identified the factors of fuel prices, fishing gear costs, ice cube prices, and unauthorized fish catch by foreign fishermen, as the other affecting factors in fish production. Concerning fisheries export, Tuna fish and prawns play a vital role in the export market. Despite the marine resources, Sri Lanka still imports fisheries products by spending foreign remittances. The Net Trade Balance (NTB) of fisheries products in Sri Lanka is investigated in this research and modelled by multiple linear regression models (net trade balance as the response variable, and harvest of 10 fish types as independent variables, based on significance) for pre and post, COVID-19 pandemic conditions. Further, the models can accurately predict the NTB (Pre-COVID model R2 = 72.4%, post-COVID model R2 = 80.6%). This model can be used in policy and strategy analysis by respective authorities such as the Ministry of Fisheries, Sri Lanka. Using the time series methods (Moving Average, Exponential Smoothing, and SARIMA), fish production is analysed. Combining these models, the impact of the COVID-19 pandemic on fish production from February to August 2020 is estimated at 15.81%. In conclusion, this research identified the fish production patterns, COVID-19's impact on production, and a model to estimate NTB, which also can be an analytical tool for the policymaking of the fisheries industry.
A quantitative characterization of sector-wise performance interdependencies in stock market using changepoint and performance-induced distant clustering
(Faculty of Science, University of Kelaniya Sri Lanka, 2023) Kavishka, R.T.; Liyanage, U. P.; Hewaarachchi, A. P.; Dissanayaka, D. M. P. V.
The Colombo Stock Exchange (CSE) is the Sri Lankan marketplace for companies to trade their stocks to the public. There are 19 sectors in the CSE before the Global Industry Classification Standard (GICS) classification. The stock market has become a significant icon in most of the country’s economy. Due to the enhancement of the data science discipline, quantitative research on the stock market has gained popularity among scholars in the recent past. Most of the studies were conducted to predict the value of a stock and its volatility. However, this study explored potential performance dependencies among the 19 industrial sectors registered at the CSE. In known literature, this scope has not been addressed quantitatively. The sectorial All Share Price Index (ASPI) is used to characterize the sector interdependencies, as the volatility of ASPI implies the sector performance at a given time or a short period. Because diverse sector movements can offset each other, leading to a stable index, while extreme sector-specific events or trends can result in increased index volatility. The ASPI indices published by CSE from 2005 to 2019 were considered in the analysis. The persistency of ASPI volatility in a compact interval indicates the consistency of the performance in each sector. Thus, the comparison of volatility changes and their changing time, i.e., changepoint analysis, describes the changes in the sectorial performances. Consequently, the interdependencies among the sector-wise performances can be recognized by the emerging patterns of the changepoints, i.e., the clustered behaviour of the changepoints. Through this approach, the investigation seeks to identify significant transitions or shifts in the behaviour of each sector. Non-parametric methods were employed in the identification of the changepoints of the ASPI series. The standard clustering approaches could not be utilized in grouping the changepoints as the clustering metrics defined by the variation of performances are interconnected. Thus, a new clustering approach was developed using a cluster performance-induced distant measure defined based on a reference industrial sector. This analysis resulted in interdependency among the industrial sectors. Further, variation patterns among the changepoints were identified using interval scaling, and dependent industrial sectors were identified with the help of the performance-induced distant clustering approach. For example, the Bank Finance & Insurance, Telecommunication, and Trading sectors exhibited strong interdependencies. Also, the Construction & Engineering, Oil palms, and Hotel & Travels sectors exhibited strong interdependencies. So far, in qualitative relations, these interdependencies were merely recognized by the gut feelings of financial analysts. Nevertheless, this study provides a clear quantitative characterization of performance dependencies, and thus, the findings are crucial for determining investment strategies and minimizing risk in stock exchanges.
Sector-wise portfolio diversification and optimization on the NYSE: adaptive clustering framework
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Wickramrathne, H. P. D. P. M.; Liyanage, U. P.
This research investigates the application of K-means cluster analysis to create diversified and optimized stock portfolios within the New York Stock Exchange (NYSE). Diversification is a crucial strategy for mitigating risk and achieving optimal returns in the stock market. Different stocks belonging to various industrial sectors exhibit diverse memory dependencies in their price profiles, particularly in volatility profiles. Therefore, general clustering approaches do not provide a stable clustering customized to each industrial sector. An advanced memory adaptive clustering method has to be utilized in order to have a stable clustering in a given industrial sector. The available methods often neglect sector-specific characteristics, leading to potentially suboptimal portfolio construction. This study proposes a novel approach that incorporates sector-wise analysis of historical data to determine the optimal number of clusters within each sector. By considering these sector-specific characteristics, the proposed methodology aims to create robust and diversified portfolios. The study uses historical closing price data for NYSE stocks from January 2020 to December 2023. To capture risk characteristics, quarterly volatility values are used as a key clustering variable. The formed clusters are then considered to represent stable risk groupings within the NYSE. Furthermore, to incorporate return potential, a similar clustering process is applied using the Sharpe Ratio, a risk-adjusted return metric. By analyzing the centroids of these return-based clusters, three distinct return levels are defined. The companies are assigned risk and return classes based on their cluster membership resulting in volatility and Sharpe Ratio analyses. This classification is used for the formulation of five portfolio strategies. Modern Portfolio Theory (MPT) is then employed to optimize a set of randomly generated portfolios for each strategy. The Sharpe Ratio serves as the optimization criterion, and the five portfolios with the highest Sharpe Ratios are selected for further analysis. The performance of these topperforming portfolios is then compared to a benchmark of ten randomly generated portfolios formed by using S&P 500-indexed companies. Evaluation based on the Sharpe Ratio demonstrates the existence of optimized cluster-based portfolios that outperform the benchmarks. In conclusion, this study offers a comprehensive framework for investors to identify and invest in companies with the potential to deliver greater profitability.
A simulation framework for investigating the market scenarios of the forex market using Ornstein-Uhlenbeck and Monte-Carlo simulation
(Faculty of Science, University of Kelaniya Sri Lanka, 2024) Kaushalya, H. A. T.; Liyanage, U. P.
The forex market is a financial market that is a global marketplace for exchanging national currencies. Forex investment is generally mediated by a broker and executed by a trader. Thereby, the investment scenarios are based on realistic exchange price fluctuations and the best market representations. Herein, all possible combinations of upward-downward movement of the exchange rate fluctuations have been considered as scenarios. However, the exchange rates have three parts namely, selling price, buying price, and absolute exchange rate, making the framework rather complex. The three price series have to be simulated focusing on the market behaviors and particular exchange rate patterns alongside price regulations undertaken by the respective authorities. As the literature suggests, in such a scenario modeling, the currency historical data analysis may not be prominent as the currency profiles have less memory dependency. Thus, in the determination of possible scenarios, memory-less stochastic processes are more appropriate than memory-depending stochastic processes. This study investigated the possibility of forming such a realistic simulation framework using appropriate mathematical tools. The exchange rates of currencies are highly volatile. The literature has illustrated that the OrnsteinUhlenbeck (OU) process shows a reliable representation of such exchange rate profiles. Thereby, this analysis focused on OU simulation representing the exchange of major nine currency pairs. The respective OU process parameters were identified based on historical data. However, parameter optimization was conducted at the level of stochastic process expectation. Thus, the Monte Carlo procedure is used in parameter estimation and has been utilized in scenario analysis as well. The respective realizations of these identified OU processes have been utilized in modeling the possible exchange rate representations and scenario simulations. As an example, the three series of USD-EURO exchange rates are simulated by the OU process with the identified parameters based on their respective historical data. A Monte-Carlo is used to formed to analyze desired strategies alongside the perspectives of brokers and traders. Using the simulation framework, we have tested the price fluctuations for traders and brokers to ensure they are realistic based on exchange rate data. The results have illustrated that brokers get the optimum profit. The profit of each process has been calculated for the broker, and the expected profit has also been calculated. Not only that, changing the parameters (mean and volatility) from 0.0001 to 0.0006 also got the expected values to check how the profits vary in that situation. After each process space was stable, identified strategies to get more profit from the broker's perspective. Finally, the strategies from the trader’s perspective were identified to get more profit while assuming there was only one trader. Also, the same process was done 1000 times and got traders an expected profit. Through the application of simulation tools, this research contributes to the current discussion about successful trading tactics in this changing environment.
Stable forecasting of tax revenues of selected countries assisted by Clustering Approach
(4th International Research Symposium on Pure and Applied Sciences, Faculty of Science, University of Kelaniya, Sri Lanka, 2019) Karunarathne, A. W. S. P.; Liyanage, U. P.; Hewaarachchi, A. P.
Tax is one of the main income of a government that utilizes in public welfare and future investment. Taxation has goals: reducing the inequalities through a policy of redistribution of income, administrating the levels of inflation as well as deflation, protecting the local industries from foreign competitions through levies, and discouraging the undesirable activities such as consumption of tobacco. Additionally, taxation provides a major portion of Gross Domestic Product (GDP), depending on the country’s fiscal policy. Tax forecasting is essential towards strategizing government plans and future activities. However, tax revenue highly fluctuates due to many factors which include natural disasters, instability of political environment and government monitory policies. This study aims to find the set of best statistical forecasting models, by comparing the behavioral similarities of different tax revenues identified by clustering approach. Here, tax revenue data from 1972 to 2017 of 24 countries belonging to developing status: developed, developing and under-developed have been analyzed. Comparable and homogenize measure is obtained considering the tax revenue as a percentage of GDP. The countries with similar tax revenue are identified by using K-Means clustering. Consequently, the selected countries were clustered into five classes depending on their tax revenue as a percentage of GDP. The analysis shows that the tax revenue has similar behavior based on the similarities of countries’ developing status. Tax revenues data in each cluster were analyzed to identify the best fitted time series models. It has been found that models of the types Autoregressive Moving Average (ARMA) and Autoregressive (AR) are best fitted models for the representing tax revenue of the corresponding clusters. As an example, ARMA (2,2) model was fitted to one cluster and AR (1) model was fitted for another cluster of countries. According to the type of the model and their range of parameter values, it is found that similar models can be used to represent the tax revenue data within the underlying cluster. That is, there exist cluster specific models in the sense of model type and their parameter ranges. This finding can be utilized towards forecasting tax revenue in the case of the revenue data are highly affected with a qualitative factor, for example, political instability. In summary, through the clustering approach, stable forecasting of revenue data of a given country can be performed.
Studying the behaviour of export quantities of Tuna fish in Sri Lanka
(4th International Research Symposium on Pure and Applied Sciences, Faculty of Science, University of Kelaniya, Sri Lanka, 2019) Sachithra, S. A. L.; Liyanage, U. P.; Wijeyaratne, W. M. D. N.
Being an island in the Indian Ocean, Sri Lanka claims a large sea area and abundant fish resource with high facilitate suitable for large scale fishery industry. According to the Central Bank of Sri Lanka, the contribution of fisheries to the Gross Domestic Production (GDP) of the country ranges between 1.3% and 1.6%. Consequently, fishery industry already plays a vital role in economics and social development of Sri Lanka. Due to weather conditions, seasonal effects, changes of government tax policies and trade agreements, e.g. GSP+ and etc., there is a high fluctuation in export quantity of fishery products in Sri Lanka. Thereby, it is essential to study the variation patterns and forecast harvest and income generated by fishery products towards monitory strategy planning. Among the various types of fish, tuna is one of the species that is important in financial earnings. Out of all fisheries exports, Sri Lanka earns the highest income worth 50.8% by exporting tuna fish in 2016, according to the statistics from Ministry of Fisheries and Aquatic Development of Sri Lanka (SLMFAD). This study was conducted to analyze the export quantities of tuna fish and forecast the future export quantities. Monthly export quantities from January, 2010 to June, 2018 were collected from SLMFAD. In preliminary analysis, United States, Japan, and Canada are identified as the top countries in which Sri Lanka exports the highest quantity of tuna fish. To study the changes in export patterns and their associated relations, Statistical Change-Point Analysis was conducted. The results revealed a high correlation between the changes of export patterns with events such as country’s peace restoration, economic stability, infrastructure facilities, introduction of different capacity changes and termination of development projects. Towards forecasting the export patterns time series data analysis techniques were used. Unit root tests; Augmented-Dickey-Fuller Test (ADF) and Kwiatkowski-Phillips-Schmidt-Shin test (KPSS) were used to test the stationarity of the time series data. Based on Akaike information criterion (AIC) value, SARIMA (1,1,2)(1,0,0)12 model was identified as the best. Ljung-Box test, Jarque-Bera test and Heteroscedasticsity test were used to check the behavior of the residuals of this fitted models. Accuracy of the models were compared by root mean squared error (RMSE), and mean squared error (MSE). With 0.8485 of RMSE and 0.6038 of MSE, SARIMA (1,1,2)(1,0,0)12 model can be considered as the most suitable model to forecast the export tuna quantity from Sri Lanka.