Please use this identifier to cite or link to this item:
Title: Predicting random walk time series and real stock prices : an experimental study
Authors: Chiew, Hong Yi
Toh, Malcolm Jia Jun
Chong, Gervais Kiat
Keywords: Social sciences::Economic theory
Issue Date: 2020
Publisher: Nanyang Technological University
Abstract: Conventional economics theories adopt the three fundamental assumptions that economic agents are fully rational, have well-defined and stable utility maximizing preferences as well as the ability to efficiently and effectively process all information. Over the past five decades, numerous studies have researched on how individuals form expectations from historical trends and series, where individuals follow either rational, extrapolative, or adaptive expectations when making their predictions. In the stock market, some individuals adopt strategies that involves examining price patterns or chasing the price trend after forming their expectations. This paper aims to investigate if individuals are inclined to engage in trending or mean reversion behaviour when viewing sequences with varying levels of reversals, autocorrelation, volatility and retreat (price range). Using OLS regression, we found that individuals tend to predict reversals when shown sequences with more reversals in sequences of random walk nature. In addition, individuals are likely to chase trends when viewing stock price movements. These results show that individuals are unable to predict the next step of a stock price as they are unable to out-predict the random walk in stock prices. Despite the random walk nature of the stock price series, overall, our participants mainly engage in both strategies after forming their expectations on the stock series.
Fulltext Permission: embargo_restricted_20220325
Fulltext Availability: With Fulltext
Appears in Collections:SSS Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
HE_1AY1920_1 FYP Main Report.pdf
  Until 2022-03-25
1.33 MBAdobe PDFUnder embargo until Mar 25, 2022

Page view(s)

Updated on Jun 13, 2021

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.