back start next
[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [ 25 ] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54]
25 directly affea the bottom line, and so these obser%ations wih become ingrained withijractice and exXiience. CALASHJaily OSf1IW2001 Mov Avg 1 line 29.288 Acme V Stralefly Linear Rl Figure 6.9. Computer Associates Volatility 7 Range Trading A iptcidator i% a . observestlieJutmv, Qiidccts before it occurs. EemardEarucli hi his 1 The Science offlitting, Ted describes how he calculated that the strike zone was approximately seven behs wide and eleen bahs high. The result was a matrix of baseballs, and he cslculated his battii aerage for each beh in the matrix. Wlule the ordinarj batter decided between beh and strike, Williams refined the strike into seentjseven separate categories [37]. Williams also determined that once a batter started switit at pitches just seeral inches outside of the strike zone, the strike zone expanded from 4.2 square feet to 5.8 square feet, an increase of almost 37 percent. Once a pitcher learned a batter would swit at bad pitches, then thats ah the batter would get, and the batter was destined to be a .250 hitter. Now, imagine if the "Splendid Splinter" apphed his analysis to ftie stock market and tumed his attention to the range. He would sort eh ofthe raises into fheh arious sizes and then determine his battit aerage, or profit factor, for each range. He would conclude that when the range is narrow (in the strike zone I his profit factor is higher. In contrast, when the range is wide (out ofthe strike zone), his profit factor is lower. The aerage trader analyzes a trade as either a winner or a losera beh or a strike. The professional trader analyzes atrade from its risk/reward ratio [13]. If the trader uses range to determine stops, then the risk numerator is the range itself (the higher the range, the higher the risk), and the reward denominator is the profit target. For example, if a long entrjis tiiggered at the high ofthe bar, and the range is 1.5 times the ATR, then the trader is probably swingii out of the strike zone. ImI W,)),.ii„. )..m<.l 1 ) ii. Ml... ii .iv. i..,.. I............ , ill. I... »li,i.ii 1>.1 1 M lilt. /.........)...m..,.> till......... I,....... .......II III ».....l..» ....I..ill ..I,.
7.1 Range Ratio The Acme N Sjstem is based on a simple concept called the Saiige Ratio. We want a ratio value less than one because a day with a low Raie Ratio (RR) is generally followed by a wide range (WR) day imder volatile maiket conditions. To calculate the RR, divide the current days range by the AverageTme Range (ATR) over a certain reference raie to estimate todays volatility. For example, ifthe ATR ofJimiper Networks forthe past se\en days is 2.5, and todays rare is 2.0, then the Range Ratio is 2.0 / 2.5, or 0.8. The Range Ratio has two irjuts. letigfhl and lengfh2. The default values are one and se\en, referred to as RR 1:7. The first range does not have to be the rare of just the current bar; it can span ammiber ofbars, so one can experiment with other ratio values such as 2:10 or 3:12. The Acme N sjstem uses a default threshold of 0.7; once the RR Mis below this vralue, the sjstem trades abreakout in the direction ofthe trend. The Raiffie Ratio indicator is a sqjarate plot that tradis the ratro ofthe ATR of one range ofbars to the ATR of another raie ofbars. When the ratio is less than a certain percentage, the chart is in consohdation and is poised to break out. When the ratio is greater than a certain percentage, then the next bar will probably be a narrow range (NR) bar. hi Figure 7.1, each time the ratio is less than 0.7 or 70°o, the next day is awide range day. iXLAST4)aJly 0 2000 MovAvs1lirte1CK>se.50,0) 10B1.9S Ac f°la="* «00.00 1150.00 1100.00 10SO,00 «50.00 800,00 liK"«7l. K....!.< 7.2 Range Patterns The Acme N sjstem integrates the Raie Ratio with other narrow range pattems de\eloped by Cooper and Crabel [4. 6]. Further, we have de\eloped other vraiiations, such as two NR5 days in a row and an NRo bar. Ah ofthese other NR pattems are part of the Acme Range Pattems, as shown in Table 7.1. TableT.l. RangePattems Pattern •rwtion ID2 Two Consecufive hiside IDNRj, hiside Dwith the Nancwest Range ofthe last n bars Two Consecutive Narrowest Range bars over n bars Narrowest Range ofthe last n bars RaiKe is x° ofthe Average True Range NR2n 7.2.1 Inside Day 2 (ID2) The Inside Day 2 pattem (ID2) is two consecutive Inside Days ( ), as shown in Figure 7.2. It is the same as Coopeis Boomer pattem [4]. rip«72. Il>2 i:)(.imi>l. I )mii„,;Jii.. •Iiiiijh.......Ik Iiim.I, 1 >.<i i. ili. Iii.l.l. \KMif, , ivhi.). In •• i.«l< .1 .>. lit
7 Trading 7.22 Inside DayNarrow Range 4 (IDNR4) The Inside DayNarrow Raie 4 pattem flDNRj) is an inside day with the narrowest range ofthe past four dajs [3], as shown inFigure 7.3. Figure7.3. IDNRiExample 7.23 Narrow Range 2 (NR2) The Narrow Raie 2 pattem (NR2) is two consecutive NR bars oer a given range. Figure 7.4 shows a chart with two consecutive NR5 dajs (NR25). Figure 7.4. .2 E:;ample 7.2 Range Pattems 7.24 Narrow Range 10 (NRio) The Narrow Raie 10 pattern (NRio) is the narrowest raie ofthe last ten dajs, as shown in Figure 7.5. F1giire7.5.NRi oExample 7.25NarrowRangeOo(NROo) The Narrow Range % pattern (NR%J is based upon a percentage ofthe ATR. Dunnigan defines an NR in the context of an Upswit or Downswh, where an NR bar is any bar with a raie less than half of the widest range bar in the swit [9]. Four examples of an NR" ojo bar are shown in Figure 7.6.
[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [ 25 ] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54]
