Comparison of Time Series Models for Forecasting Sales in Multi Level Marketing (MLM) Business

Abstract

This study aims to analyze and compare the performance of time series models in forecasting sales for network marketing businesses. Traditional models such as ARIMA are compared with modern models like Facebook Prophet. The study uses actual monthly sales data from network marketers, covering the period from January 2025 to February 2025, totaling 1.5 months. The performance of the models is evaluated using statistical metrics including RMSE, MAE, MAPE, and R² Score. The results reveal that ARIMA tends to provide more accurate forecasts than Facebook Prophet. This research is important for the network marketing sector, as accurate sales forecasting enables entrepreneurs to effectively plan marketing strategies and manage inventory. Furthermore, comparing the effectiveness of time series models supports business owners in selecting suitable tools to analyze sales trends and business growth.

However, the research is still in the experimental phase, and results may change as more diverse data becomes available.