Forecasting the Number of Complaints with the Office of the Consumer Protection Board

Abstract

The Office of the Consumer Protection Board (OCPB) collects statistical data on consumer protection and analyzes trends and directions of consumer protection problems in various dimensions in order to supervise and follow up on concrete solutions to consumer protection problems. However, no advance predictions have been found yet. Changing trade patterns from traditional commerce to the e-commerce era and social media advertising that reaches consumers through smart devices will cause trade transactions to increase, resulting in an increase in the number of complaints submitted to the Office of the Consumer Protection Board (OCPB). This article aims to forecast the number of complaints for fiscal year 2023 using a time series model based on Gray System Theory, which can be used well with small amounts of data, for use with the number of complaints from fiscal year 2018 to 2022. The forecast is divided into four time series: real estate and housing, consumer products and services, and the total of the three groups above. The prediction results of the GM (1,1) expanded with periodic correction model (EPC) had the least Mean Absolute Percentage Error (MAPE), so the EPC model was chosen. It is predicted that fiscal year 2023 will have a number of complaints that increased or decreased from fiscal year 2022, as follows: Real estate group: 2,131 cases, a decrease of only 0.46 percent; consumer products group: 8,689 cases, an increase of 12.93 percent; service group: 6,193 cases, a decrease of 6.50 percent; and overall 17,244 cases, an increase of 4.78 percent. When combining the forecast values from the 3 groups, there will be a forecast value for the number of cases in fiscal year 2023 of 17,012 cases, which is only 232 cases less than the overall forecast of 17,244 cases, a decrease of only 1.34 percent.