Personal Financial Forecasting Model

Built a few forecasting models to determine the demand for a particular product.

My thesis proyect about NN to forecast time series

Exploring forecasting for energy consumption

Forecasting model construction for custumer volume prediccion per each hour in two stores

This project aims to forecast an exchange rate for dollars/GBP on August 8, 2020, using a Neural Network model in SPSS.

Predicting amoebiasis cases in Southern Taiwan using machine learning methods and transfer entropy-based feature engineering

Detect U.S. housing market bubbles using macroeconomic signals. Forecast HPI, score speculative risk, and visualize insights using a fully modular, cloud-native GCP pipeline.

Forest the unit sales of Walmart retail goods for the next 28 days

My exercise-focused version of the primary repository for the course: Data Science, Prediction, and Forecasting, taught as part of the CogSci masters program at Aarhus University.

Inflation forecasting pipeline combining ARIMA diagnostics in Stata with LSTM tuning in Python, featuring dynamic forecasts and evaluation metrics (MSE, MAE, R²).

Using machine learning to forecast engagement of trending songs on the TikTok social media platform. Should marketing strategies change to maximize user engagement?

Forecasting Bitcoin Trend Using Sentiment Extraction From Twitter

Solutions to real world machine learning problems

Repository of numerical models for pseudo-forecasting of induced and natural seismicity in the Hengill geothermal field. Developed in the framework of the COSEISMIQ project http://www.coseismiq.ethz.ch/en/home/ by Antonio P. Rinaldi, Shyam Nandan and Vanille A. Ritz

Source code for the coastal inundation forecast demonstration project for Fiji

Time Series Forecast of traffic on JetRail

The goal of this project is to investigate whether intra-day data can improve the performance of LSTM models in daily time series forecasting tasks.

This project aims to forecast the stock price of Bank Negara Indonesia (BBNI) from 2012 to 2030 using multiple time series forecasting techniques.