Machine Learning Model for Sales Forecasting by Using XGBoost

Abstract

For modern retail corporations operating a huge chain of businesses, exact sales predication is the key in driving corporations development, even success or failure. Sales forecasting allows corporations to efficiently allocate resources including cash flow, production, and make better informed business plan. In this paper, we propose an efficient and accurate sales forecasting model using machine learning. Initially, feature engineering is conducted for extracting features from historical sales data. Furthermore, we used eXtreme Gradient Boosting (XGBoost) to utilize these features for forecasting the future sales amount. The experiment results on a publicly Walmart retail goods dataset provide by Kaggle competition demonstrate our proposed model performs extremely well for sales prediction with less computing time and memory resources.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 85 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.