Cement-Stock-level-Prediction

The main objective of this project is to predict the stock demand according to previous cement sales data. The company practices a traditional prediction system using spreadsheets which makes it difficult to deal with the big data. And the accuracy of sales forecasting is challenging due to the volatile and non-linear nature of the financial stock demand. Since stock demand forecasting is crucial in any sector, it has recently gained huge popularity to boost market operations and productivity due to new technologies. Therefore, we implemented a time-series prediction to forecast future stock demand. The project attempts to implement four machine learning algorithms including Radial Basis Function (RBF), Multivariate Regression, Random Forest and Long Short Term Memory(LSTM). We compared the performance of these algorithms and tested their accuracy.

Methodology

1. Data Collection and Exploration

2. Data Preparation and Preprocessing

3. Algorithm Implementation

   3.1 Radial Basis Function (RBF)

   3.2 Multivariate Regression

   3.3 Random Forest Regression

   3.4 LSTM(Lib)

4. Model Building

5. Results

Prepared by

1. Sintayehu Zekarias
2. Fikir Awoke
3. Ashenafi Kifleyohans