Option Investor
Trader's Corner

Time Frames in Trading Decisions; 'DMI' Indicator

Printer friendly version

In response to a SUBSCRIBER E-MAIL I was prompted to consider: "Your recent three articles on moving averages, indicators and oscillators made me look at, evaluate and test some of my screening, entry and exit parameters. Still to soon to tell what changes I want to make but I'm liking what I'm seeing.

So here's my next question. For short term/swing trading, what time periods (Example, 1 year, 6 month, 20 day, 5 day, 1 day?) do you recommend for:

1. Identifying trend, support, resistance, chart patterns, etc.
2. Determining entry points based on directional or momentum strategy.
3. Monitoring the trade and determining trade progress and exit points."

Your question may be a bit too broad or my answer too general for me to nail down what would be a helpful response, but you can always send a follow up e-mail on this, as needed to clarify. You'll note that I tend to use 2, 3, 5, 8, 13, 21 as key time frames, as they're part of the 'fibonacci' number progression.

I tend to take the meaning of 'swing trading' as attempting to capture and profit from the short-term trend, which I define as the 2-3-5 day price swings, although a short-term trend can extend to 8-10 trading sessions.

For me, the short-term trend in practical terms? It's basically the length of time that a 21-day hourly RSI will tend to go from 'overbought' (65-70) to oversold' (35-30) or vice versa, in a market that is experiencing moderate to strong price momentum.

I don't tend to use a moving average on hourly charts, but do use 21 as the 'length' setting for the hourly oscillators like RSI. On daily charts I use the 21-day moving average and 21-day moving average envelopes as most helpful for timing intermediate (2-3 week) price swings. I assume trade ABOVE the 21-day average will tend to have more upside potential than not, and trade BELOW the 21-day average more downside than not.

I use a 13-day RSI most often on daily charts, but like the 21-day stochastic for the Dow Jones Average.

The 50 or 55-day moving average and the 200-day moving average are helpful in terms of seeing the long-term trend. Trade above the 200-day moving average is a good gauge of a long-term uptrend and vice versa as to a downtrend.

On the weekly chart, I tend toward the equivalent 10 and 40-week moving averages. I use an 8, sometimes, 13-week RSI to judge overbought/oversold on a weekly chart basis. MACD is helpful on weekly charts at the 12-26-9 default settings on my TradeStation, which is what the inventor of the study used.

In terms of time, I rely on hourly charts the most to define the short-term trend and look the most at the last 2-5 days worth of data. For the intermediate-term trend of 2-3 week price swings I use daily charts. For the longer-term 2-3 month trend I also use daily charts, but also look at weekly charts.

I look at the intermediate trend, in terms of trading OPTIONS, as the 2-3 week price swings. The 'long-term' trend for me is the 2-3 month dominant trend.

I don't look at definitions of trend, support/resistance, and chart patterns as having particular time duration. I might start looking at prior lows or highs as support and resistance for the past 3-5-8 days then broaden that out to look for prior lows or highs as potential support and resistance for the past 2-3 weeks. That is as far as prior highs or lows.

As far as trendlines, the relevant factor is using the steepest up or down trendline that has 2-3 lows or highs with which to define it. Then look at the next trendline with a less steep slope or angle.

Resistance has been seen such as at the recent top when OEX reached a daily chart trendline drawn through prior highs on the daily S&P 100 chart below. Recent lows found support in the area of prior highs; prior resistance, once broken, 'became' support later on. This thinking was supported by the support found in the area of the 21-day moving average. There was one close under the average yesterday, but as occurred at the early-March low, there was not more than a 1-day close under this key trading average.

A swing trader might buy today's higher opening, with an exit point just under the prior day's low. A minor down trendline is drawn from the top that must be pierced (at 592) to suggest that OEX was going to re-test the previously mentioned trendline drawn through the prior highs and intersecting currently around 597.

The 21-day moving average envelope line, set to float 2.5% above the average and at 605 currently, is where OEX would begin to be 'extended' on the upside, relative to its normal trading fluctuations above/below this average in recent weeks/months.

As with moving averages, I tend to wait for a second consecutive day's close above or below a trendline to 'confirm' a break. An example being the ONE daily close seen in February at the long-term up trendline on the chart above and the low for that move.

An overbought reading in the 65-70 range on the 13-day RSI above was seen in the daily chart and occurred close to the onset of the current short-term (it appears) downside correction. As is typical of up trends, past months downswing lows didn't reach the 'fully' oversold reading typically seen at 35-30 in my RSI.

The hourly OEX chart, which is a 'close up' view of the unfolding chart pattern below, showed a well defined uptrend price channel WITHIN which there were shorter-term term trading swings, with two taking place over 5-6 trading days and one of around 10 days. Hourly trendline breaks are useful in deciding to EXIT a profitable trade. I rely a lot on hourly charts for monitoring trades.

However, within the February 10-day downtrend, there were two trading swings of 3-5 days that a very short-term trader could have captured. Use of 30-minute charts rather than 60 (hourly) can be helpful in trading for shorter durations and objectives.
Use of a 21-period RSI on 30-minute charts remains useful also.

What characterized all the below short-term trends, was that there were fairly well defined up or down trendlines that developed in the course of the Index move. When a trendline got pierced, especially after a reversal at a BIGGER duration trendline (e.g., one measuring 30-60 day durations), the subsequent move was worth trading. In hindsight the below highlighted patterns look easy to define; NOT so easy is figuring out where one trend ends and a (trend) reversal begins WHEN it's happening. But, trendlines do provide a major help in trading in a large number of instances of making swing-trading decisions.

There was not always a higher or lower extreme seen on the RSI that touched the upper or lower lines as I have them set current (35,65), but the tops were at the upper extreme and at the lows the 21-hour RSI approached the lower extreme.

In an intermediate to long-term uptrend a general rule of thumb is to expect more times when an Index or individual stock gets to a fully 'overbought' extreme before there's a short-term top. In a downtrend, expect fewer times for a 'fully' oversold reading to be reached.


"I would like to see Trader's Corner address about the DMI buy signal; which people commonly use in Charts with some variable like as: ADX (14) 22.... This signal is reliable for Buy?"

Welles Wilder developed the Directional Movement Index (DMI) and he applied it to the commodity futures markets, which is what he traded. The DMI attempts to provide an indication of how much 'directional movement' (trend) is present and provides a way to compare trends in different markets. You can see that this might be of particular use in commodities markets, which are not as well correlated as stocks and indexes within the stock market. As well, there are more PROLONGED straight up or straight down price swings in commodities driven as they are by supply and demand; e.g., imagine what a freeze does to the coffee or orange crop.

There are 3 component lines to this Indicator and the DMI is seen on the S&P 500 (SPX) chart below. The ADX line (the cyan 'histogram' that looks like the outline of a mountain below) rates the directional movement on a 1 to 100 scale. The HIGHER the ADX line, the MORE that the market is trending and the better candidate it is for use in a 'trend following' system. A very low ADX line at or below 20, indicates a non-trending environment, which has been the case since early this year.

TTwo other lines are generated in the DMI Indicator, DMI+ (green line) and DMI- (the red line). DMI+ measures 'positive' or upward movement and DMI- measures 'negative' or downward movement. A buy 'signal' is given when the DMI+ line crosses ABOVE the DMI- line. A sell 'signal' is when the DMI+ line crosses BELOW the DMI- line. Buy/sell 'signals', but not the first buy in the yellow circle, are noted with up green arrows (buy) or red down arrows (sell). 14 is set as 'length'; this can be any number you wish.

As seems to be suggested in your question, you could decide to take ONLY the buy (DMI+ crosses above DMI-) 'signals' because you see the intermediate to long-term trend as UP. That improves the result in the chart shown above. It should be noted that this Indicator, like Moving Averages, is a 'LAGGING' Indicator. Therefore, as with all lagging indicators, it takes a period of time for rising prices to trigger an upside crossover and a similar period of time for a downside crossover in a downtrend. br>
Use of this Indicator in a strictly 'mechanical' way, will tend to lead to WHIPSAW trading 'signals'; i.e., by the time that a buy signal is given (except for the FIRST one), there is not a big further move before there is an opposite signal. Especially of course in a non-trending or 'trading range' market like we've seen since late last year. HOWEVER, here's where ADX comes in.

Trading results can be dramatically improved by taking ONLY the buy/sell signals when the ADX is ABOVE 20 (a trending market) and not taking crossover type signals with the ADX either falling (see chart above) and especially not with ADX at and below 20.

On this basis the DMI might only suggest taking a handful of trades in a year. This would be OK if you have the discipline to WAIT for just the right conditions. And why not! Not if you care only about ending the year at a decent profit!

Please send any technical and Index-related questions for answer in Trader's Corner articles to support@optioninvestor.com with 'Leigh Stevens' in the Subject line.

This Wed. Trader's Corner article also serves as an adjunct to my weekend 'Index Trader' column. In the article you're reading here I can briefly update a technical picture of the market as of midweek, and then use recent chart/indicator patterns to more fully explain their relevance to trading decisions in general.

More on my specific predictions, support and resistance, etc. is found in my weekend Index Trader column, available on the Option Investor.com WEB site (not part of the e-mailed weekend OI Daily). You will normally see a web LINK to the Index Trader at the top of your weekend Option Investor Daily e-mail, but in the rare event that the link goes missing (as happened this past weekend), my Index Trader can still be found on the OptionInvestor.com web site. My most recent (Sat, 3/25) Index Trader can be seen online by clicking here.

** Good Trading Success! **

Trader's Corner Archives