Stock price forecasting remains a prominent area of research in financial markets due to the complicated, nonlinear, and volatile nature of stock price paths. While high-performance deep learning models like Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU) demonstrate superiority in financial forecasting, their comparative efficacy in emerging markets, such as the KSE-100 Index, remains inadequately explored. This study examines the prediction performance of these models based on past stock prices of the KSE-100 Index and based on standard performance measures (root mean square error, mean absolute error, and mean absolute percentage error) to provide a holistic assessment of the models. Our research indicates that GRU consistently performs better than CNN and LSTM, with lower errors and higher predictive accuracy. This work enhances current forecasting methods by utilizing advanced data preparation, feature engineering, and hyperparameter optimization to augment model performance. The results have practical implications for policymakers and investors in emerging markets by emphasizing the applicability of deep learning to financial decision-making.
In the technique of paired comparisons, objects are ranked since individual judgment. We use it when quantifiable measurement is not viable or impractical. In this study, the Glenn-David PC model under Bayesian framework is used to establish the rating of five brands of cold drinks. Bayesian analysis has been made using non-informative and informative priors. The posterior means are considered for the preference behavior of the cold drink brands. The predictive probabilities for a single future paired comparison of cold drink brands are also found. The posterior probabilities of the hypotheses for comparison of parameters for any two cold drink brands are obtained. Also, the preference probabilities for paired comparison are determined. The results obtained through Uniform and Normal-Gamma priors are compared. It is observed that similar results and same rankings for the cold drinks brands are achieved. The appropriateness of the model is tested by chi-squared statistic. All computations are made in SAS package by designing the programs/codes.
In the prevention of heart disease, the identification of risk factors plays a pivotal role. For this purpose, the present study aims to identify the lifestyle and health factors that contribute to heart disease in individuals aged 26 to 45. The goal of this study is to identify the risk factors like personal, clinical, and lifestyle that significantly increase the prevalence of coronary heart disease in adult males and females aged 26 to 45 years. Data is collected from the Preventive Department of the Punjab Institute of Cardiology (PIC) in Lahore, Pakistan. In 2021, from August 9 to September 22, we collected data from 300 participants through surveys and medical records, half of whom had cardiac disease, and the other half were healthy. Factors such as high blood pressure, smoking, stress, exercise habits, cholesterol levels, weight, age, family history, and gender were also taken into consideration. Using the binary logistic regression model, it was revealed that the OR (Pvalue) of smoking 3.898 (0.000), high blood pressure 2.598 (0.001), a family history of heart disease 2.433 (0.001) and being 41–45 years old 4.162 (0.007) were all significant risk factors for developing coronary heart disease. To decrease heart disease in Pakistan, smoking prevention, high blood pressure treatment, cholesterol maintenance, and family history awareness should be significantly focused.
Being economical, RMDs are used in the experiments of almost every field of life. When RMDs are applied, there are chances to rise the residual effects. Residual effects can be removed through washout periods, but washout periods cannot be applied in many situations. To balance the residual effects, minimal circular balanced RMDs are preferred which can only be constructed for odd v = ip+1 through method of cyclic shifts (Rule I). Partially balanced RMDs-II can be used for odd v = ip+3 which are not available in literature. In this article, therefore, these designs are constructed using Rule I for 3 ≤ p ≤10. Efficiency measures of residual effects and of Separability for each design are also calculated.
For the unspecified population mean estimation, a family of exponential type estimators has been proposed when survey encounters measurement error under simple random sampling without replacement. Up to the first approximation degree, the mean square error and bias expressions of suggested estimators have been derived. Numerical and simulation studies were done to validate and compare the estimator with some other existing similar type estimators by using R.
