Search
× Search
Tuesday, December 24, 2024

Archived Discussions

Recent member discussions

The Algorithmic Traders' Association prides itself on providing a forum for the publication and dissemination of its members' white papers, research, reflections, works in progress, and other contributions. Please Note that archive searches and some of our members' publications are reserved for members only, so please log in or sign up to gain the most from our members' contributions.

Still Talking Strategy Makeup

photo

 Guy R. Fleury, Independent Computer Software Professional

 Tuesday, February 9, 2016

Recently, I've chronicled in this forum, a stock trading experiment which kind of got lost with the new layout. I had not provided an analysis of the outcome. My objective, in this experiment, was to show that you could take an ordinary trading script and transform it into a portfolio builder. I considered the task worthwhile. It is also why I opted to do it live, posting the results as I went along. A way of forcing me to look for better solutions and be more creative. So, to remedy the situation, here is part of the post-analysis: http://alphapowertrading.com/index.php/papers/196-a-stock-trading-strategy-experiment-v Hope some will find it interesting, and maybe, most importantly, useful.


Print

3 comments on article "Still Talking Strategy Makeup"

photo

 Stephane Hardy, Computational Finance Quant and Options Trader

 Thursday, February 11, 2016



Please use events and not time series path estimations. If a wolf is running after his prey, the path of the prey is irrelevant statistically before the prey notices he is in danger. After that, the envelope of his escape choices is limited by the wolf's and the prey's dynamic constraints. No statistics here. Bayesian responses an parameter memories comme into play only as limits to the position of the target. Hence we do not have a continuous hypothesis on the position of market prices on the time and sales data feed. Generally the transition probability has no information. The next price is the current price. Unless your observations and data manipulations, identify and trigger an event. Thanks. Stephen.


photo

 Guy R. Fleury, Independent Computer Software Professional

 Friday, February 12, 2016



@Stephane, I think you have a misunderstanding of what is at work in that strategy, or, I really don't understand what you are talking about.

The MACD is an after event delayed trade triggering mechanism. It was demonstrated in the first part of this series that it had very little, if not none at all, in predictive powers.

It's not: “Please use events and not time series path estimations...”. The events, whatever they may be going forward, all have their respective impact incorporated in a MACD which is a simple moving average crossover thingy. And if there is very little in predictive powers, the notion of making “time series path estimations” becomes not only totally irrelevant but a trivial pursuit leading to an estimated total profit, from all the trading, tending to something close to zero, as was shown in part 1 of the series.

...more


photo

 Guy R. Fleury, Independent Computer Software Professional

 Friday, February 12, 2016



You add: “...The next price is the current price.” Think about it!

That is the definition of a mathematical martingale, and for it to hold requires that the expected value of all future prices has little if not no deviation from the initial price however long the time series. So, I do not adhere to such a notion. Look at any price chart, and it will tell you that: the next price can be more than significantly different from the current price. Stock prices are not flat lines, as your statement declare and affirm. Should your statement be true, then E[Δp] = 0 obliterating any trading strategy and rendering them totally useless. It is not only on a MACD based trading strategy, but all of them, since their long term profit expectation would tend to zero. There is no need to trade whatsoever if there is no profit expectation.

In payoff matrix notation, you get: Σ(H.*E[ΔP]) → 0.

Please login or register to post comments.

TRADING FUTURES AND OPTIONS INVOLVES SUBSTANTIAL RISK OF LOSS AND IS NOT SUITABLE FOR ALL INVESTORS
Terms Of UsePrivacy StatementCopyright 2018 Algorithmic Traders Association