Department of Statistics
Permanent URI for this communityhttps://repository.cuilahore.edu.pk/handle/123456789/25
Browse
7 results
Search Results
Item INTERNSHIP AT AUR-LAB(Library Information Services COMSATS University Islamabad Lahore Campus, 2024-04-02) FAZEEL UMAR; FA20-BST-002; Dr. Mian Muhammad FarooqAUR Lab is a cooperative venture between Predominant College, as of now positioned No. 1 private for influence among private colleges in Pakistan (QS Positioning), and We-Plan, an Australian plan firm working in Pakistan. In the span of 10 months of foundation, it has accomplished the achievement of one of its startup projects being shortlisted among top two out of roughly 100+ entries from Pakistan in the Endeavor Cup Global contest. Current undertakings are managing in various business ventures, including archeological the travel industry; Metaverse; NFT Commercial center; administration and casting a ballot; Style plan; healthcare; Shrewd Agrarian; environmental the travel industry; structural plan; AR-based promoting; energy effectiveness; street wellbeing; orientation equity; also, land income. Our group of specialists incorporates specialized staff as well as business improvement staff: Dr. Muhammad Ashraf Khan; Prof. Saleem Zubair; Ms. Qurratulain Sonia Kashmiri; Mr. Muhmmad Ahmad; Mr. Jawad Ahmad; also, Mr. Javaid IqbalItem Spatial and Temporal Analysis of Drought(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Muhammad Furqan Ahmad; CIIT/FA19-BST-002/LHR; Dr. Mian Muhammad Farooq; LHR TP 9937In the recent years climate change is the hot topic of research. Climate change can have severe effect on the environment. The drought is an important factor of the climate. If a drought occur in some area it will affect the agriculture, water resources and other important factors of the country. In this study, the impact of drought in different districts of Punjab, Pakistan was studied using Standardized Precipitation Index (SPI) and Standardized Anomaly Index (SAI). The SPI is applied on the rain data and the SAI was applied on the temperature data. The data was collected from Pakistan Metrological Department. At first, we fit the data on Length Biased Exponential Distribution and then we applied Standardized Precipitation Index on the rain data received from Pakistan Meteorological Department from 1993 to 2022. We also computed SPI using 3-month, 6-month, and 12-month moving average of the original data. We conducted spatial and temporal analysis using SPI and for the visual representation we also plot maps of the past five years for better understanding of the drought. Standardized Anomaly Index (SAI) is a useful tool to measure and monitor deviations from normal conditions in terms of temperature or other relevant variables associated with drought. We used temperature to better explain the drought impact in Pakistan. After applying SAI, we find out that temperate of the data is above average which means there is drought impact in different districts of Pakistan. The results show the clear drought pattern in different districts of Pakistan. We ignore the 0 and negative values of rain data before fitting the Length Biased Exponential Distribution as it is undefined for these values and then we compute the results. The graphs show that after taking moving average the drought patterns decrease in districts and also few years does not show any drought pattern.Item Modified Radial Basis Function Network (RBFN) by Bayesian Regression for predicting Synthetic Cancer(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Fazeel Umar; CIIT/FA20-BST-002/LHR; Dr. Mian Muhammad Farooq; LHR TP 9948This thesis investigates the use of Long Short-Term Memory (LSTM) networks to predict Google’s stock prices. The study focuses on stock data from January 2012 to December 2016 for training, and January 2017 for testing. LSTM, a type of Recurrent Neural Network (RNN), is ideal for time series forecasting because it can learn long-term dependencies. To prepare the data, stock prices were normalized using Min Max Scaler, which helps improve model performance. The data was then organized into sequences of 60-time steps using a sliding window approach. The LSTM model was built with four layers, each containing 50 units, and included dropout layers to reduce overfitting. Training was conducted using the Adam optimizer and mean squared error as the loss function over 100 epochs with a batch size of 32. The results showed that the LSTMmodel effectively captured the stock price patterns, highlighting its potential for accurate financial forecasting.Item RNN for Time series Forecasting Using Google Stock Prices(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Sharjeel Ahmed; CIIT/FA20-BST-011/LHR; Dr. Mian Muhammad Farooq; LHR TP 9955This thesis investigates the use of Long Short-Term Memory (LSTM) networks to predict Google’s stock prices. The study focuses on stock data from January 2012 to December 2016 for training, and January 2017 for testing. LSTM, a type of Recurrent Neural Network (RNN), is ideal for time series forecasting because it can learn long-term dependencies. To prepare the data, stock prices were normalized using Min Max Scaler, which helps improve model performance. The data was then organized into sequences of 60-time steps using a sliding window approach. The LSTM model was built with four layers, each containing 50 units, and included dropout layers to reduce overfitting. Training was conducted using the Adam optimizer and mean squared error as the loss function over 100 epochs with a batch size of 32. The results showed that the LSTMmodel effectively captured the stock price patterns, highlighting its potential for accurate financial forecasting.Item Classification of heart disease using MLP with SeLU activation function and Binary Cross Entropy as Loss function(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Nouman Ali; CIIT/FA20-BST-009/LHR; Dr. Mian Muhammad Farooq; LHR TP 9954Multilayer perceptron are the basic components of deep learning and neural net- work characterized by their architecture of feedforwading consisting of fully con- nected neurons with the non linear activation functions.This thesis provides a comprehensive overview of MLPs,by giving detailed about their structure and com- ponents,including the input layer,hidden layers, and output layer. This explains the role of weights and bias neurons in the process of learning and adjustment of the complex pattern in data. Different activation functions which are Sigmoid, ReLU, tanh, were used, illustrating their important in non-linearity and using the model to find its relationship.The training of MLPs by using the back propagation method is thoroughly studied. This includes the forward pass,where the input data is stored and processed though the network,and the backward pass, where gradients of loss function are calculated to update the network parameters. The importance of the loss estimation and parameter updates are used in minimiz- ing the error and improving the model’s performance.Additionally, we will discuss about the important steps in data preparation, which involves the handling missing values,feature scaling, and ensuring the proper data formatting. The importance of splitting the data into training and testing is highlighted, Technique for preventing the over fitting, such as regularization,dropout, and early stopping. The guidelines for using the MLP is provided, starting with the simple one and then increasing the complexity based on different tasks.This thesis focuses on the usage of exper- imentation with various architecture,hyper parameters and optimizing the model to get the optimal results. Different techniques used for monitoring the training and the adjustment of hyper parameters are discussed, along with the methods to calculate the model accuracy and loss checking and test datasets.This thesis aims to adopt practitioners with the understanding of tools to develop robust and efficiency of model on different applications such as image recognition,NLP, and speech recognition. By following the detailed guidelines and using the best tech- niques in data preparation, training and evaluation can increase the performance of model.Item Transmuted Marshall Olkin Modified Burr-III Distribution(Library Information Services, COMSATS University Islamabad, Lahore Campus., 2020-04-01) Ali Zain; CIIT/FA19-BST-004/LHR; Dr. Mian Muhammad Farooq; LHR TP 9939Transmutation of classical probability distributions play a significant role in real life modeling. In this work, we develop a new Transmuted Marshall Olkin Modified Burr-III distribution. The proposed model is named Transmuted Marshall Olkin Modified Burr-III distribution (TMOMBIIID). Several statistical characteristics of TMOMBIIID such as survival function and hazard rate, are discussed. The Maximum likelihood method is followed for the estimation of model parameters.Item Generalized Inverted Exponential Distribution: Theory and Applications(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Muhammad Umar Amin; CIIT/FA22-RST-005/LHR; Dr. Mian Muhammad Farooq; LHR TP 9871In this research work we introduced generalization of the Inverted Exponential Distribution based on survival function. Our aim is to provide an extensive treatment of the mathematical properties of new model. The possible extension of corresponding probability density and distribution function have been derived and present. Some distributional properties of new model are discussed and order statistics have been derived. Maximum likelihood estimation (MLE) is sued to estimate the model parameters and real-life data application have also used to study the proposed model. Additionally, the study uses Bayesian estimation techniques, obtaining posterior estimates of the model parameters through the use of the Markov Chain Monte Carlo (MCMC) procedure. This method offers more accurate inference than the standard methods. Furthermore, Type-II censoring is used to evaluate the model performance, indicating its flexibility and reliability.