A Machine Learning Approach to Forecasting Consumer Food Prices
Date
2017-08-24T14:05:44Z
Authors
Harris, Jabez (Jay)
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Building on the success of the Canada Food Price Report 2017 and its inclusion of a machine learning methodology, this research thesis posed and attempted to answer the following question, “What is the best way to predict food prices for the average Canadian consumer?” The Canada Consumer Price Index (CPI) was selected as the dependent variable and forecasted against three data models to access their predictive values. The models included the popular Holt-Winters Triple Smoothing Exponent model as a benchmark, a financial futures-market data model and a model adapted from the Canada Food Price Report 2017. The hope was to create a more robust forecast model for future Canada Food Price Reports and similar econometric predictions. As hypothesized, the Financial Futures-Market based model outperformed the Food Price Report model with a 1.6% average error rate.
Description
This research thesis attempted to answer the following question, “What is the best way to predict food prices for the average Canadian consumer?” The overall objective was to forecast the Canada Consumer Price Index (CPI) to assess the performance of various machine learning techniques against three data models. Specifically, the research aimed to: 1. Determine the top performing model of the three models assessed (Holt-Winters, Food Price Report, Financial Futures-Market) 2. Determine the top performing machine learning technique of the four assessed (Linear Regression, Multilayer Perceptron, SMOreg, M5P Tree) 3. Evaluate the performance of the Multilayer Perceptron as the only technique to incorporate backpropagation. This research thesis was not meant to be a definitive approach to food price forecasts by favoring one technique over another but rather was intended to illustrate the accuracy of machine learning techniques in this forecasting domain by using the tools and techniques which can easily be duplicated by the average consumer.
Keywords
Machine Learning, Food Price, Neural Network, Futures Market, Multilayer Perceptron, Holt-Winters, Backpropagation, Backtesting