Sökning: "movement prediction"
Visar resultat 1 - 5 av 74 uppsatser innehållade orden movement prediction.
1. On modelling OMXS30 stocks - comparison between ARMA models and neural networks
Master-uppsats, Uppsala universitet/Matematiska institutionenSammanfattning : This thesis compares the results of the performance of the statistical Autoregressive integrated moving average (ARIMA) model and the neural network Long short-term model (LSTM) on a data set, which represents a market index. Both models are used to predict monthly, daily, and minute close prices of the OMX Stockholm 30 Index. LÄS MER
2. A Markovian Approach to Financial Market Forecasting
Kandidat-uppsats, KTH/Matematisk statistikSammanfattning : This thesis aims to investigate the feasibility of using a Markovian approach toforecast short-term stock market movements. To assist traders in making soundtrading decisions, this study proposes a Markovian model using a selection ofthe latest closing prices. LÄS MER
3. Electron-skyrmion systems, in and out of equilibrium, and isolated or contacted to reservoirs
Magister-uppsats, Lunds universitet/Matematisk fysik; Lunds universitet/Fysiska institutionenSammanfattning : A Kondo lattice skyrmion model in contact with a macroscopic environment is simulated to explore skyrmion dynamics, which is an extension of previous work. The system is simulated using non-equilibrium Green's functions within the generalized Kadanoff-Baym ansatz and the wide band limit. LÄS MER
4. Improving Water Droplet Prediction for Vehicle Exterior Water Management: Insights from Experimental and Simulation Studies
Master-uppsats, KTH/Väg- och spårfordon samt konceptuell fordonsdesignSammanfattning : This thesis focuses on the study of water transportation on vehicle surfaces, which is crucial for ensuring the unobstructed operation of sensors and cameras in autonomous vehicles. The research aims to develop and validate experimental and simulation methods to enhance the understanding of water droplet behaviour and to create accurate models for computational fluid dynamics (CFD) simulations. LÄS MER
5. Temporal Localization of Representations in Recurrent Neural Networks
Master-uppsats, Högskolan Dalarna/Institutionen för information och teknikSammanfattning : Recurrent Neural Networks (RNNs) are pivotal in deep learning for time series prediction, but they suffer from 'exploding values' and 'gradient decay,' particularly when learning temporally distant interactions. Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) have addressed these issues to an extent, but the precise mitigating mechanisms remain unclear. LÄS MER