Review: Detecting Outliers with Distributions for Estimating Time Series Models. | ||||
| The Egyptian Statistical Journal | ||||
| Article 2, Volume 68, Issue 2, December 2024, Page 24-33 PDF (490.87 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/esju.2024.306371.1037 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Mohamed Ezzat Abd-Elaziz  1; Ahmed Amin El-Sheikh2; Amal Mohamed Abd-Elfatah2
| ||||
| 1Masters student | ||||
| 2Department of Applied Statistics, Faculty of Graduate Studies for Statistical Research, Cairo University, Giza, Egypt. | ||||
| Abstract | ||||
| In statistical analysis, outliers represent data points that significantly deviate from the general pattern of a dataset. Understanding and addressing outliers is crucial because they can skew results, impacting the reliability and validity of conclusions drawn from data analysis. This paper provides an in-depth exploration of outliers across three key dimensions. First, it offers a general overview of outliers, discussing their characteristics, methods, and algorithms used to detect them in various contexts, including fields such as finance, healthcare, and social sciences. Second, it examines outliers within distributions, detailing how they influence measures such as mean, median, variance, and standard deviation, and how they can affect the overall shape and interpretation of the data distribution. Techniques for detecting and mitigating the impact of outliers are also discussed. Third, it analyzes outliers within time series data, focusing on their potential to distort trends, cyclic patterns, and forecasting accuracy. By investigating these dimensions, this paper aims to enhance the understanding of outliers and underscore their significance and challenges in statistical analysis. | ||||
| Keywords | ||||
| Outlier; Time series; ARIMA; Autoregressive; Exponential; Gamma; Additive outlier | ||||
|
Statistics Article View: 283 PDF Download: 211 |
||||
