Kaggle Sales Prediction

1. Problem Statement

As a business analyst for a global e-commerce company, you’re always looking for innovative ways to improve sales strategies and streamline operations. The company has recently ventured into selling limited-edition Kaggle-branded merchandise, including stickers, to foster a stronger sense of community among its users. The management team wants to better understand the dynamics behind sticker sales across their international stores and improve their inventory management and promotional planning.

You have been given several years of historical sales data from their various online stores, which is in a tabular format. Your mission is to explore the data to identify seasonality and other trends, then build a model that accurately forecasts sales, enabling the company to optimize its inventory levels, schedule targeted promotions, and ultimately increase profitability. This model will help to avoid overstocking or stockouts, ensure efficient resource allocation, and better meet customer demand, contributing to the company’s overall success and enhancing its relationship with the Kaggle community.

Goal: The goal of this project is to build a model that can accurately predict sales for various Kaggle-branded stickers in different fictitious stores across different countries.

2. Data Description

The dataset contains sales data for various Kaggle-branded stickers from different fictitious stores across different countries. The dataset is completely synthetic but contains many effects you see in real-world data, e.g., weekend and holiday effect, seasonality, etc.

• Rows: 230131
• Columns: 6
• Variables:
  • id: Unique identifier
  • date: Date
  • country: Country
  • store: Store
  • product: Product
  • num_sold: Number sold (target variable)

Data Source: Kaggle Playground Series S5E1

3. Your Task

Your task is to build a model to predict sales for various Kaggle-branded stickers.

1. Data Exploration and Preprocessing:
   • Load the dataset using Pandas.
   • Explore the data to understand the distribution of features.
   • Handle missing values, if any.
   • Convert the date column to appropriate format.
2. Feature Engineering (Crucial):
   • Create new features that may improve model performance (e.g., day of the week, month, year).
3. Model Building:
   • Split the data into training and validation sets.
   • Choose a suitable model (e.g., Time Series, Random Forest, Gradient Boosting).
   • Train the model on the training data.
4. Model Evaluation:
   • Evaluate the model's performance on the validation data.
   • Use metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE).
5. Prediction and Submission:
   • Make predictions on the test data.

Python Libraries: Pandas, NumPy, scikit-learn, Matplotlib/Seaborn.