ARIMA-Based Prediction for the Number of People Reporting to Wordle

Abstract

As soon as Wordle is launched, people all over the world are posting their scores on Twitter with great enthusiasm. The projection of the number of people helps the development of the software. To test the applicability of the ARIMA model in this area. In this paper, the data is fed into the model for prediction. The analyses show that the value of R2 is 0.982, so the model fits well. The predicted values also show that the number of people reporting scores has a downward trend. This paper divides the attributes of the word into six dimensions. We use the Least Squares Estimation, Spearman Correlation Analysis and ADF Test to analyze the relationship between six attributes and the percentage of scores reported that were played in Hard Mode. The test shows that there was neither linear nor non-linear relationship between them.