Time Series Forecasting in Stock Trading Markets

Edward J Lusk

doi:10.20525/ijrbs.v8i4.283

Abstract

Univariate Time Series Models [TSM] use only a Panel of historical data to produce forecasts. The tacit belief in using TSM is that the past information portends the future of the longitudinal data-stream. This is likely in certain cases such as strictly Ergodic Panel segments of sufficient size in the overall Panel. A question of interest is: Is the success of TSM in these contexts generalizable? The test of this question used a Litmus-Test design to examine the performance profile of TSM for a longitudinal time series the last point of which is a Turning Point [TP]. Specifically, the inference measure will use the Relative Absolute Error [RAE] of the TSM tested over three forecasting horizons. In this testing, five TSM configurations were employed; the TPs are identified using a fixed screening filter applied to randomly selected firm Panels actively traded on the S&P500 from 2005 through 2013. There is no evidence that any of the five TSM outperformed the RW model which is incidentally the TP. The impact of these results is that one cannot assume that the effectiveness of TSM generalizes to all domains—in particular—forecasting after TPs that seems to be a Domain Lacuna where the effectiveness of TSM will be compromised.

Highlights

This introduction first offers a socio-Human Information Processing perspective to rationalize the development of the forecasting context addressed in the research
Lusk (2019b) notes in a paper that creates forecasting enhancements for use in the BloombergTM terminals, in particular, the FA platforms that: Archimedes remarked: Give Me a Fulcrum, and I Shall Move the World; the simile in the market trading world is; Give Me an Effective Forecasting Model and I can make Bill Gates and Sam Walton look like Paupers
The intention of offering a brief historical perspective on the “Need to forecast Accurately and the commensurate Rewards” and to introduce the TSM group of forecasting models sets the following context for this research report: As the desire to find an accurate forecasting model seems a part of our gnome and TSM are the simplest class of effective forecasting models perchance forecasters are lead to believe the TSM class to be the Forecasting Grail—i.e., they are the universal key to unlocking the future

Summary

Introduction

This introduction first offers a socio-Human Information Processing perspective to rationalize the development of the forecasting context addressed in the research. The Magic 8 Ball is owned by Mattel, Inc., and as of 2012 still sold more than a million units per year (https://www.britannica.com/story/where-did-the-idea-forthe-magic-8-ball-come-from) These vignettes offer evidence of the human pre-occupation with believing, usually against the common sense of experiential reality, that forecasters that appear to be in a trance-like-state [whether induced by noxious fumes or self-induced as the necessary artistic accoutrement to enable the suspension of dis-belief] provide glimpses into worlds that have yet to happen. If this is the case, and it certainly seems to be, forecasting models which use real mathematical and statistical platforms and operate in the real Big-Data stock market context are even more likely to garner “devotees” as they offer a dimension of Holdback reality checking and the lure of monetary rewards. Peer-reviewed Academic Journal published by SSBFNET with respect to copyright holders

Literature Review

Design of the TSM Test

Summary of the Above Testing

Findings

Conclusions