https://github.com/ramyacp14/salesforecasting
Forecasts future sales for a retail company using time series analysis with Facebook Prophet. The project involves data preprocessing, exploratory data analysis (EDA), and forecasting with holiday effects considered.
https://github.com/ramyacp14/salesforecasting
data-preprocessing data-visualization exploratory-data-analysis fbprophet machine-learning matplotlib model-evaluation numpy pandas python seaborn time-series-forecasting
Last synced: about 1 month ago
JSON representation
Forecasts future sales for a retail company using time series analysis with Facebook Prophet. The project involves data preprocessing, exploratory data analysis (EDA), and forecasting with holiday effects considered.
- Host: GitHub
- URL: https://github.com/ramyacp14/salesforecasting
- Owner: ramyacp14
- Created: 2022-12-18T04:17:49.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2024-09-06T19:56:01.000Z (over 1 year ago)
- Last Synced: 2025-01-13T08:46:23.417Z (about 1 year ago)
- Topics: data-preprocessing, data-visualization, exploratory-data-analysis, fbprophet, machine-learning, matplotlib, model-evaluation, numpy, pandas, python, seaborn, time-series-forecasting
- Language: Jupyter Notebook
- Homepage:
- Size: 1.73 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: Readme.md
Awesome Lists containing this project
README
# Sales Prediction Project
This project focuses on predicting future sales for a retail company using time series analysis and the Facebook Prophet library.
## Project Overview
The project analyzes sales data from multiple stores and aims to forecast future sales. It includes data preprocessing, exploratory data analysis, and time series forecasting using Facebook Prophet.
## Data Sources
The project uses two main datasets:
1. `train.csv`: Contains daily sales data for multiple stores
2. `store.csv`: Contains additional information about each store
## Key Features
- Data cleaning and preprocessing
- Exploratory Data Analysis (EDA) with visualizations
- Time series forecasting using Facebook Prophet
- Incorporation of holiday effects in the forecast
## Requirements
- Python 3.6+
- pandas
- numpy
- matplotlib
- seaborn
- fbprophet
## Usage
1. Mount your Google Drive (if using Google Colab)
2. Import the required libraries
3. Load and preprocess the data
4. Perform EDA
5. Train the Prophet model
6. Make predictions and visualize results
## Main Functions
- `sales_prediction(Store_ID, sales_df, holidays, periods)`: Generates sales predictions for a specific store
## Future Improvements
- Fine-tune the Prophet model parameters
- Incorporate additional external factors that might affect sales
- Implement cross-validation for model evaluation
- Extend the analysis to include all stores in the dataset