Algeria's CPI Inflation Forecast Using DNN Methods
Data Analysis | Data Science |Python | Tensorflow | Keras | Streamlit | Numpy | Pandas.
Data Analysis | Data Science |Python | Tensorflow | Keras | Streamlit | Numpy | Pandas.
The accurate forecasting of Consumer Price Index (CPI) inflation is a crucial aspect of economic planning and decision-making. In the context of Algeria, where economic stability and development are paramount, the need for precise inflation predictions becomes even more pronounced. Traditional forecasting methods such as ARIMA may not fully capture the intricate patterns and dynamics inherent in inflation time series data. Therefore, there is a compelling need to explore and leverage advanced deep learning models, such as Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Convolutional Neural Network (CNN), to enhance the accuracy and reliability of CPI inflation forecasts.
Exploratory Data Analysis (EDA):
Conduct a comprehensive Exploratory Data Analysis (EDA) on the CPI inflation rate, utilizing visualizations such as heatmaps, boxplots, autocorrelation plots, and seasonal decomposition.
Provide insights into the key characteristics and patterns of the CPI inflation series, aiding in a better understanding of the data.
Train DNN Models:
Implement the training process for each model (MLP, RNN, LSTM, BiLSTM, CNN) using TensorFlow and Keras in Python.
Assess the performance of diverse deep learning models (MLP, RNN, LSTM, BiLSTM, CNN) in forecasting Algeria's Consumer Price Index (CPI) inflation.
Conduct a comparative analysis to identify strengths and weaknesses, guiding the selection of the most effective models based on forecasting accuracy and their suitability for the specific characteristics of Algerian inflation data.
Interactive Forecasting Application:
Develop an interactive forecasting application using Streamlit, allowing users to explore and visualize CPI inflation forecasts dynamically.
Implement features that enable users to switch between different periods and models, enhancing the application's usability and accessibility.
This project's significance lies in its potential to influence economic decision-making, empower businesses and investors, enhance public awareness, contribute to technological advancements, and further academic research in the field of economic forecasting. By bridging the gap between advanced machine learning techniques and economic indicators, the project aligns with the broader goal of fostering economic resilience and sustainability.
We begin with the unadjusted Algerian consumer price index (CPI), in the period January 2000 to December 2022. The series was gathered from IFS-IMF. We transform the adjusted series to a monthly year-on-year consumer price inflation rate.