Ask Colin

Do you need to change the number of periods for a 200-day moving average if changing from a daily to a weekly chart?

If I use a certain moving average on weekly data in my chart, do I need to change the period of the moving average if I look at daily data? For eg, is a 200-day moving average on a weekly data chart valid also on the daily data chart? Or if I switch from the weekly to the daily, do I need to change the MA period as well? I am quite confused about this and haven’t been able to find a solution anywhere.

I have actually addressed this issue in a couple of answers to  questions on my website on the Free Resources menu/Ask Colin Page under the Moving Average keyword.

However, here is another answer that may be more technically complete:

It depends  on how the charting software you are using calculates moving averages. There are two basic ways different charting software does it:

  1. Some software always calculates moving averages based on the daily data. So, the display (daily, weekly or monthly bars) has no impact on the calculation of the moving average, which in your question would always be a 200-period moving average. Thus, the moving average line will be identical on the daily, weekly and monthly bar charts – the bars are different periods, but the moving average is always 200-periods.

  2. Some charting software compresses the underlying data for both the display and the calculation of indicators when you draw a weekly or a monthly bar chart. So, with this kind of software (Insight Trader, which I use does it this way), you must adjust the number of periods:

  • For a 200-day moving average on a daily bar chart, use 200 periods for the moving average

  • For a 200-day moving average on a weekly bar chart, you must use 40 periods for the moving average (200 days divided by five days in a week)

  • For a 200-day moving average on a monthly bar chart, you must use approximately 9 periods (200 days divided by 21.67 days per month)

With the second type of software, since the actual closing prices used (200 x daily, 40 x weekly or 9 x monthly) are different, the three moving averages will be slightly different. I myself use 260 days or 52 weeks or 12 months. For my purposes as an investor there is no significant difference. However, if you are trading very short term, where small price differences are significant, I suggest constructing all three on a dozen or so charts that seem fairly different in volatility. Then study whether the differences are significant for your method. If your software allows it, draw the three charts over the same number of years and overlay the three charts, suppressing the bar charts so you have just the three moving average lines. This will expose how much different they are.

I hope this fully addresses your question.

Keyword: