Predictive analytics is the use of data, statistical algorithms and machine learning techniques to identify the likelihood of future outcomes by using current and historical information. Over the last 15 years John MacLeod has researched and applied these methods to the ASX Stock Market to underpin his private trading. Recent research coupled with learning out of the ATAA and member collaborations has further enhanced the predictive power of this modelling through blending TA indicators with regression and support vector machine (SVM) predictive methods.
Trading will always be conducted at some point in the future, so it makes sense to estimate the likely future change of a stock price at future points in time and this is where predictive analytics comes into play. Furthermore, providing a measure of confidence with this prediction is even more useful. John's presentation is intended to prompt his audience to think about:
- using TA attributes in predictive models that estimate the likely future change in price of each stock;
- measuring the strength of each stock estimation;
- how all stocks within a constantly changing market will all have distinct intrinsic drivers price movement behaviours; and
- why models should be constructed and executed in order to accommodate constantly changing markets
John MacLeod has a Master of Science in Physics and 30 years in a professional analyst & consulting background around designing, building and implementing Predictive and Trigger Based analytical models for telco's, local and overseas consumer banking, publishers, FMCG and other mass marketing organisations.
In 2004 John commenced investigating various methods for predicting the direction of stock prices in the ASX in order to run and support his SMSF and this has resulted in the successful ongoing management of his SMSF. Over the last 3 years since joining the ATAA and Quant Trader Special Interest Group along with collaborating research with other ATAA members these predictive methods have been much enhanced.
This presentation will challenge the way we use TA and will introduce predictive analytics and some data science as an alternative way for finding trades in the market.