All Actions In The Markets Amount To Forecasts

All actions in the markets based on any method, technical or fundamental, amount to forecasts. Any claims that technical or fundamental analysis are not used to make forecasts but for risk management purposes only are naive and demonstrate ignorance.

Technical and fundamental analysis can be used to identify market entries, exits and to manage risk. In all these cases, any actions to establish new positions, exit any open positions or increase/decrease risk exposure amount to making a forecast about future market direction. It is easy to see this, both logically and quantitatively.

The logic

Apparently, there is no reason for any action based on any method unless the trader believes that profits will be made or losses will be minimized. This is a forecast irrespectively of method used. By claiming of just reducing or increasing risk, the fact is that a forecast is involved and that cannot be concealed. Obviously, both fundamental and technical analysis can be used for risk management, this is not disputed here. The point is that anyone who claims that no forecasting is involved is naive and actually unaware of the essence of any actions that amount to forecasts. If there is no forecast, the only reasonable action is doing nothing. If action is taken, it can only be justified in the context of some forecast, implicit or explicit. Even in passive indexing and allocation, there is forecasting involved that the choice will result in positive expectation.

The math

This was proven already in QUSMA blog in response to well-known trend-follower who claimed that he just follows markets and not forecasting anything. The proof is simple and should remove any doubts. Below I discuss another mathematical proof.

Suppose there is an open long position that was established based on some method, technical or fundamental, or even randomly, with N number of shares or contracts. The entry price is Pe.  The profit/loss P/L from the position is given as follows

P/L = (Px – Pe) × N × BPV

where Pe is the exit price and BPV is the big point value. In the case of stocks, BPV = 1.

Next, suppose that a trader uses technical or fundamental analysis to either determine Pe or change N by buying or selling shares/contracts to increase/decrease risk.

In fact, any collection of returns based on entry points Pe, exit points Px and position size Nk as a function of trade number k is a forecasting model since it results in an average trade Tavg equal to expectation E[T] for sufficient samples.  Therefore, for every action taken there is an associated expectation. This is forecasting. QED.

But why making naive claims?

Some people, especially those who have fallen behind in the era of machine learning, algo and quantitative trading, try to distant themselves from this new environment by claiming that their use of technical analysis is only for risk management and even that their trading is not probabilistic.

To start with, any claim of non-probabilistic trading is extremely naive. Although probabilities are difficult to calculate and in some cases unknown, trading is probabilistic because it deals with a stochastic system where same causes produce different effects depending on prevailing market conditions.

As shown above, although technical analysis is valuable for risk management, as I also argue in my book Fooled By Technical Analysis, each and every action based on it, or even on fundamental analysis, involves a forecast because there is an implicit or explicit (calculated) expectation that by virtue of the action some objective will be reached, for example maximizing profit, limiting losses, etc.

If you have any questions or comments, happy to connect on Twitter: @mikeharrisNY

Charting and backtesting program: Amibroker

Disclaimer

Technical and quantitative analysis of Dow-30 stocks and 30 popular ETFs is included in our Weekly Premium Report. Market signals for longer-term traders are offered by our premium Market Signals service. Mean-reversion signals for short-term SPY traders are provided in our Mean Reversion report.