Browsing by Author "Peiris, T.S.G."

Now showing 1 - 6 of 6

An Application of 5-fold Cross Validation on a Binary Logistic Regression Model
(2016) Attanayake, A.M.C.H; Jayasundara, D.D.M.; Peiris, T.S.G.
Abstract Internal validation techniques can be used to check the predictive ability of the developed models. The most common internal validation techniques are split sample methods, cross validation methods and bootstrapping methods. The split sample methods are inefficient with the small size of data sets. The bootstrapping methods are efficient with the knowledge of computer programming languages. The cross validation methods are not very popular in practice. Therefore, in this study 5-fold cross validation method of cross validation techniques is applied to validate the predictive ability of a binary logistic regression model. The binary logistic regression model was fitted on a data set of UCI machine learning repository. Results of the cross validation reveal that low value of optimism and high value of c-statistic in the fitted regression model indicate an acceptable discrimination power of the developed model.
Impact of mathematics on academic performance of engineering students: A canonical correlation Analysis
(Faculty of Science, University of Kelaniya, Sri Lanka, 2016) Nanayakkara, K.A.D.S.A.; Peiris, T.S.G.
Mathematics plays a key role in higher education as it is particularly essential to develop the analytical thinking of students. Mathematical skills would certainly assist to enhance students’ knowledge in a wide range of disciplines, especially, in engineering sciences. Therefore, exploring the student academic performance has received great attention among researchers recently. The main objective of this study is to investigate the impact of mathematics on students’ academic performance at the end of Level 2, in different engineering programs. The study was conducted with engineering undergraduates from seven different disciplines at the Faculty of Engineering, University of Moratuwa, Sri Lanka in academic year 2011/2012. Students’ examination marks of mathematics courses in Level 1 and Level 2 and all compulsory engineering courses in Level 2 were used for the study. Explanatory data analysis techniques and canonical correlation analysis were used to achieve the objectives. Statistical testing confirmed that only the first canonical function is significant for all engineering disciplines. The amount of variance between the students’ performance in mathematics and engineering courses in Level 2 explained is varied from 39% to 73%. The students’ performance in engineering courses in both semesters of Level 2 is positively and strongly related to mathematics performance irrespective of the engineering disciplines. Furthermore, the combined effects of mathematics in Level 1 and Level 2 on students’ performance in engineering courses in Level 2 are significantly higher compared with the individual effect of mathematics in Level 1 or Level 2. The combined effects of mathematics in both Level 1 and Level 2 are immensely beneficial to improve the overall academic performance at the end of Level 2 of the engineering students. However, the impact of mathematics varies among engineering disciplines. The students are encouraged to achieve high marks in mathematics courses for better performance in engineering courses.
Measuring Stock Market Volatility in an Emerging Economy
(2011) Peiris, T.U.I.; Peiris, T.S.G.
The pattern of volatility in a given time series is due to various micro and macro economic factors attached to that security. An understanding of volatility and its causes is important in determining the cost of capital of the security and in assessing investment and leverage decisions in case of emerging economies especially where the market consists of risk?averse investor. This study thus examines the volatility of different sectors in Colombo Stock Exchange (CSE) and how the macro economic factors affect on the volatility by fitting Autoregressive Conditional Heteroskedasticity (ARCH) and the Generalized ARCH (GARCH) using monthly time series data of 20 sectors in CSE for the period 2005-2010. Results found that sixteen out of twenty sectors in CSE has a significance volatile (p<0.05) and both ARCH and GARCH terms on the fitted models for individual sectors were significant (p<0.05). The volatility of composite stock returns of volatile sectors was then regressed against Narrow Money Supply (M1), Broad Money Supply (M2), Inflation (I) and Interest Rate (IR). It was found that inflation and interest rate are the two significantly influencing macro economic factors (p<0.05) on the stock market volatility of emerging economies like Sri Lanka.
Statistical Analysis of Road Traffic Accidents (RTAs) in Sri Lanka
(Department of Social Statistics, Faculty of Social Sciences, University of Kelaniya Sri Lanka, 2021) Kodithuwakku, D.S.; Peiris, T.S.G.
Road Traffic Accidents (RTAs) are one of the most prominent public health problems as it is a leading cause of death by injury and all deaths globally. This study, therefore, intended to determine the significant factors associated with RTAs in Sri Lanka (2005 - 2019) and the impact of those factors using data obtained from the Department of Police, Sri Lanka. The leading causes for RTAs are overtaking, speed driving and diversion and about 80% RTAs are due to these factors. The percentage of RTAs due to alcohol consumption by the driver is around 9%. Both exploratory and confirmatory factors analysis found that the causes for RTAs can be classified into two independent factors namely, (i) negligence of pedestrians & drivers and (ii) lack of attention of the driver. These factors are invariant by the factor extraction method and the type of orthogonal rotation. The condition of road the surface, light condition of the road, the situation of weather, type of vehicle and age of the driver are significantly influential factors in fatal accidents. The highest percentage of fatal accidents have occurred when the road is wet and light condition is poor during night. The inferences derived from this study can be effectively used for policy decisions related to traffic in order to minimize RTAs in Sri Lanka. The study confirmed the benefits of data-driven decision-making for policy decision process.
Validation of the World Health Organization/ International Society of Hypertension (WHO/ISH) cardiovascular risk predictions in Sri Lankans based on findings from a prospective cohort study
(Public Library of Science, 2021) Thulani, U.B.; Mettananda, K.C.D.; Warnakulasuriya, D.T.D.; Peiris, T.S.G.; Kasturiratne, K.T.A.A.; Ranawaka, U.K.; Chakrewarthy, S.; Dassanayake, A.S.; Kurukulasooriya, S.A.F.; Niriella, M.A.; de Silva, S.T.; Pathmeswaran, A.; Kato, N.; de Silva, H.J.; Wickremasinghe, A.R.
INTRODUCTION AND OBJECTIVES: There are no cardiovascular (CV) risk prediction models for Sri Lankans. Different risk prediction models not validated for Sri Lankans are being used to predict CV risk of Sri Lankans. We validated the WHO/ISH (SEAR-B) risk prediction charts prospectively in a population-based cohort of Sri Lankans. METHOD: We selected 40-64 year-old participants from the Ragama Medical Officer of Health (MOH) area in 2007 by stratified random sampling and followed them up for 10 years. Ten-year risk predictions of a fatal/non-fatal cardiovascular event (CVE) in 2007 were calculated using WHO/ISH (SEAR-B) charts with and without cholesterol. The CVEs that occurred from 2007-2017 were ascertained. Risk predictions in 2007 were validated against observed CVEs in 2017. RESULTS: Of 2517 participants, the mean age was 53.7 year (SD: 6.7) and 1132 (45%) were males. Using WHO/ISH chart with cholesterol, the percentages of subjects with a 10-year CV risk <10%, 10-19%, 20%-29%, 30-39%, ≥40% were 80.7%, 9.9%, 3.8%, 2.5% and 3.1%, respectively. 142 non-fatal and 73 fatal CVEs were observed during follow-up. Among the cohort, 9.4% were predicted of having a CV risk ≥20% and 8.6% CVEs were observed in the risk category. CVEs were within the predictions of WHO/ISH charts with and without cholesterol in both high (≥20%) and low(<20%) risk males, but only in low(<20%) risk females. The predictions of WHO/ISH charts, with-and without-cholesterol were in agreement in 81% of subjects (ĸ = 0.429; p<0.001). CONCLUSIONS: WHO/ISH (SEAR B) risk prediction charts with-and without-cholesterol may be used in Sri Lanka. Risk charts are more predictive in males than in females and for lower-risk categories. The predictions when stratifying into 2 categories, low risk (<20%) and high risk (≥20%), are more appropriate in clinical practice.
Validation of the World Health Organization/ International Society of Hypertension (WHO/ISH) cardiovascular risk predictions in Sri Lankans based on findings from a prospective cohort study
(Ceylon College of Physicians, 2020) Thulani, U.B.; Mettananda, K.C.D.; Warnakulasuriya, D.T.D.; Peiris, T.S.G.; Kasturiratne, K.T.A.A.; Ranawaka, U.K.; Chackrewarthy, S.; Dassanayake, A.S.; Kurukulasooriya, S.A.F.; Niriella, M.A.; de Silva, S.T.; Pathmeswaran, A.P.; Kato, N.; de Silva, H.J.; Wickremasinghe, A.R.
INTRODUCTION AND OBJECTIVES: There are no cardiovascular(CV)-risk prediction models specifically for Sri Lankans. Different risk prediction models not validated among Sri Lankans are being used to predict CV-risk of Sri Lankans. We validated the WHO/ISH (SEAR-B) risk prediction charts prospectively in a population-based cohort of Sri Lankans. METHOD: We selected participants between 40-64 years, by stratified random sampling of the Ragama Medical Officer of Health area in 2007 and followed them up for 10-years. Risk predictions for 10-years were calculated using WHO/ISH (SEAR-B) charts with- and without-cholesterol in 2007. We identified all new-onset cardiovascular events(CVE) from 2007-2017 by interviewing participants and perusing medical-records/death-certificates in 2017. We validated the risk predictions against observed CVEs. RESULTS: Baseline cohort consisted of 2517 participants (males 1132 (45%), mean age 53.7 (SD: 6.7 years). We observed 215 (8.6%) CVEs over 10-years. WHO/ISH (SEAR B) charts with and without-cholesterol predicted 9.3% (235/2517) and 4.2% (106/2517) to be of high CV-risk ≥20%), respectively. Risk predictions of both WHO/ISH (SEAR B) charts with- and without-cholesterol were in agreement in 2033/2517 (80.3%). Risk predictions of WHO/ISH (SEAR B) charts with and with out-cholesterol were in agreement with observed CVE percentages among all except in high risk females predicted by WHO/ISH (SEAR B) chart with-cholesterol (observed risk 15.3% (95% Cl 12.5 - 18.2%) and predicted risk 2::20%). CONCLUSIONS: WHO/ISH (SEAR B) risk charts provide good 10-year CV-risk predictions for Sri Lankans. The predictions of the two charts, with and without-cholesterol, appear to be in agreement but the chart with-cholesterol seems to be more predictive than the chart without-cholesterol. Risk charts are more predictive in males than in females. The predictive accuracy was best when stratified into two categories; low (<20%) and high (≥20%) risk.