Outliers are data points or observations that stand out significantly from the rest of the group in terms of size or
frequency. They are also referred to as "abnormal data". Before fitting a forecasting model, outliers are often
eliminated from the data set, or if not removed, the forecasting model is altered to account for the presence of
outliers. The first scenario covered in the study is the detection of outliers when the parameters have been
established. Second, where there are unidentified parameters. This article mentions a number of causes for outlier
correction and detection in time series analysis and forecasting. For the objective of the study, a time series of the
volume of water entering the Dohuk dam reservoir in Dohuk city was used. The study arrived at the following
conclusions after conducting their research: first, whenever the critical value increased, the value of residual
standard error (with outlier adjustment) increased. Second, the quantity of outlier values dropped each time the
critical value was raised. Third, forecasts with outlier correction perform better than forecasts without outlier
adjustment when outliers are present.
See More
See Less