Explainable Machine Learning Models and Architectures

10Revenue Forecasting Using Machine Learning Models

Yashasvi Roy and Sanmukh Kaur*

Amity School of Engineering and Technology, Amity University Uttar Pradesh, Noida, India

Abstract

Machine learning (ML) has come a long way and has a significant impact on how we analyze data nowadays. In this chapter, we consider financial data of an organization with different variables and categories. All the categorical data has been encoded so that an ML algorithm can comprehend it. The data is run through different ML algorithms to find the one suitable or best fit for the data and providing maximum accuracy. Different performance metrics have been employed to determine the best fitting model. It has been observed that gradient boosting algorithm and a combination of multiple ML algorithms provide the best results for financial forecasting.

Keywords: Forecasting, encoding, machine learning, quantitative forecasting

10.1 Introduction

In this work we analyze forecasting of a company’s revenue. It is a technique of making financial predictions. The organization’s state is ultimately determined by its sales figures.

Financial forecasting also requires accurate predictions of future fixed and variable costs as well as various sources of income. On the basis of previous performance data, predictions are created. These assist in predicting future trends.

Businesses and entrepreneurs use financial forecasting to determine how to spend their resources or what the estimated costs for a given time ...