Rapid identification of outstanding real estate investment trusts with outlier detection algorithms

Citation

Ng, Keng Hoong and Khor, Kok Chin (2016) Rapid identification of outstanding real estate investment trusts with outlier detection algorithms. Journal of Theoretical and Applied Information Technology, 88 (2). pp. 321-330. ISSN 1992-8645

[img] Text
Rapid identification of outstanding real estate investment trusts with outlier detection algorithms.pdf
Restricted to Repository staff only

Download (486kB)

Abstract

Finding outstanding stocks is always the primary goal of an investor. This is because outstanding stocks tend to outperform others in investment return. However, uncover this type of stocks from a stock pool requires extensive financial knowledge and consistent efforts in analyzing the abundant amount of financial data. Thus, it is impractical for an amateur investor. The objective of this study is to rapidly identify outstanding stocks from the Real Estate Investment Trust (REIT) sector. We adopted two outlier detection algorithms, i.e. Interquartile Range (IQR) and Local Outlier Factor (LOF) to trace REIT stocks that were deviated from the average performers. Subsequently, the outstanding REIT stocks can be identified from the small amount of outliers. The entire process is speedy and can be done on the fly. The identified outstanding stocks were assessed based on their 1-year average total return as compared with the non-outlier stocks. The preliminary result showed that their average total return is better than its non-outlier peers.

Item Type: Article
Uncontrolled Keywords: Financial ratio, Interquartile range, Local outlier factor, Outlier stocks, Real estate investment trusts
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 20 Feb 2017 04:47
Last Modified: 20 Feb 2017 04:47
URII: http://shdl.mmu.edu.my/id/eprint/6460

Downloads

Downloads per month over past year

View ItemEdit (login required)