Tag Archives: Summary statistics

A Comparative Time Series Analysis of Points Scored by Arsenal Football Club in Premiership Competitions Using Descriptive and Probability Modeling (Stochastic) Approaches (Published)

This paper discusses the performance of Arsenal Football club London from 1893/1894 to 2013/2014 seasons by considering the points scored every season. The method of analysis adopted includes comparing the forecasting strength of two time series analysis approach: Descriptive and Probability modelling approaches. Summary statistics which includes three accuracy measures (Mean Absolute Deviation (MAD), Mean Square Deviation (MSD) and Mean Absolute Percentage Error (MAPE)) of the residuals and forecast errors from the fitted quadratic models and Autoregressive Integrated Moving Average (ARIMA) (2, 2, 0) models were used for the comparison. The data was assessed for variance stability and it was found that there is need for square root transformation. The forecasting strengths of the two approaches were compared using the forecast errors of the two models. The results of the analyses showed that the descriptive model performed better than the fitted probability model.

Keywords: Buys-Ballot table, Data transformation, Summary statistics, group averages and standard deviation and Differencing