Timing the Market with the Golden Cross

One of the most common techniques to timing the market (and timing the entry/exit in individual positions) is a moving average crossover system.

What is a moving average (MA) crossover system?

A moving average crossover system generates buy and sell signals when the moving averages of a price cross. The shorter period moving average responds to changes in price faster than the longer period moving average. Applying a crossover system to a basket of stocks or a market index creates market timing.

The faster MA crossing the slower MA from below implies the market is favorable. When the faster MA crosses the slower MA from above, it implies the market is adverse. For a stock market timing system, favorable means enter the market and adverse means go to cash (or short the market).

NOTE: If you are unfamiliar with moving averages and crossover systems, see the Chart School at StockCharts.com.

Golden Cross parameters

A default moving average periods for the Golden Cross are 50 day and 200 day simple moving averages (SMAs) of the daily close or 10 week and 40 week SMAs of the weekly close. You can use either since both daily and weekly historical data is available from Yahoo! historical quotes.

How effective is the system?

It really depends where you look on the timeline. If you build out the system in a spreadsheet or TA tool like MetaStock or AmiBroker, you will be able to pick periods when timing the market outperforms and periods when buy-and-hold outperforms. Maximum drawdown is also a measure of system performance, not just equity growth.

Roll your own stock market timing system

Go to Yahoo Finance and download the historical price data for yourself if you do not have it in an existing TA tool.

NOTE: You need to do the crossover with data that has not been changed to reflect the dividends. You need to do the performance with data that includes dividends to get total return.

Load the data into a spreadsheet or your favorite Technical Analysis back-testing system and have at it.

Keep in mind it is pretty easy to spend significant amounts of time optimizing a data set for performance. The results may be useless unless you have followed some sound back-testing and system building rules. If you torture the data long enough, it will confess to anything. Does this sound like the voice of experience?

- Reminder -
"Past performance is no guarantee of future results".

Return to Timing the Market

Popular Pages

CANSLIM Strategy
Market Timing
Portfolio123 Screener
Ranking Systems


We value your suggestions, comments, and questions.

Our goal is to make this site as useful as possible.


Contact Us

Thanks!


Consistent, sustainable investing processes.

About Me


Why build a website instead of a blog?

About this site