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]


31

Smooth(4), BuyStop( . ), Short5top(0.0);

ATR = AverageCRange, Length); BuyStop = High + (EntryFactor * ATR); Shortstop = Low - (EntryFactor * ATR);

If DataCompression < 2 Then

CoffmentaryCThis indicator must be applied to a daily b;

"interval or longer.") Else Begin

PatternString - "";

LongString = "";

ShortString = "";

ATR = Volatility(Length);

PriceDelta = ATR / 4;

I-f Close of Data2 > 0 Then Begin PatternString - "W;

HighLow = AcmeHighLowIndexfClose of Data2, Length); If HighLow = 1 Then

LongString = LongString + PatternString Else If HighLow = 2 Then

ShortString = ShortStiing + PattemStrii;

End;

If Close of Data3 > 0 Then Begin PatternString = "P";

HighLou = AcmeHighLowIndeK(Close of Oata3, Length); If HighLow = 1 Then

LongString = LongString + PatternString Else If HighLow = 2 Then

ShortString = ShortString + PatternString;

End;

If Close of Data4 > 0 Then Begin PatternString = "H";

HighLow - AcmeHighLowIndex(Close of Data4, Length); If HighLow = 2 Then

LongString = LongString + PatternString Else If HighLow = 1 Then

ShortString = ShortString + PatternString;

End;

If dose of Data5 > 0 Then Begin PatternString = "L";

HighLow = AcmeHighLowIndex(Close of DataS. Length); If HighLow - 1 Then

tongString LonRString i PatternString FIm If Hi[lnw ? Ttirn

I ill): I I.ii iiiiiM I ill)::

......

If Close of Data6 > 0 Then Begin PatternString = "V; HighLow = AcineHlghLoiandex(Ave Length);

je(Close of Data6, Smooth),

If HighLow = 1 Then

LongString = LongString + PatternString Else If HighLow = 2 Then

ShortString = ShortString + PatternString;

End;

If Close of Data? > Then Begin PatternString = "B";

HighLow = AcineHighLowIndex(Close of Data?, Length); If HighLow = 2 Then

LongString = LongString + PatternString Else If HighLow = 1 Then

ShortString = ShortString + PatternString;

End;

If Close of DataB > 0 Then Begin PatternString = "S";

HighLow = AcineHighLowIndex(Close of DataS, Length): If HighLow = 1 Then

LongString = LongString + PatternString Else If HighLow = 2 Then

ShortString = ShortString + PatternString;

End;

{Calculate shares based on risk model}

N = AcmeGetShares(Equity, RiskModel, RiskPercent, RiskATR); {Multiple Pattern Buy Signal}

If StrLen(LongString) >= MinimumPatteins Then Begin

(Draw Entry Targets on the Chart}

If DrawTargets Then

Conditioni = AcfneEntrylargetsCH", BuyStop, 0, 0, 0);

BuyCAcme LE narket") N Shares Next Bar on BuyStop Stop; End;

{Multiple Pattern SeU Signal)

If StrLen(ShortString) >= HininumPatterns Then Begin {Draw Entry Targets on the Chart} If DrawTargets Then

Conditioni = AcmeEntryTargetsCM", 0, o. Shortstop, o); SellCAcme SE Market") N Shares Next Bar on Shortstop Stop; End; End;



i Market Models

8.4 Exan4>les

8.4 Examples

Figure 8.12 displaj-s the chart ofthe S&P 500 index from September 2001 to March 2002. The third bar in the chart shows a bullish WS" bar (VIX and Short Sales Ratio), but a long entrj was not tiiered. The frfth bar shows a bullish "VHLB" bar with the following confirmations:

a High Confirmation for the VTX

a Low Confirmation for New Highs

a High Confirmation for New Lows

a High Confirmation for the Bullish Consensus

Along entrjwas triggered the first week of October, and the maiket rallied lOo o-er the next two months. Shortly after the rally stalled in Januarj of 2002, a bearish "VHLBS" bar occurred with the following confLrmations:

a Low Confirmation for the VIX

a Higli Confirmation for New Highs

a Low Confirmation for New Lows

a Low Confirmation for the Bullish Consensus

a Low Confirmation for the Short Sales Ratio

A short entrj was triggered the following week with a subsequent 7° decline.

(SPXLASHWeelily 03/01/2002 Acme Market MDdel(2.2D)

I m LEMamet

11200.00 11SO.O0

iieo.oo

LT J-114O.O0 1120.00 1180 0 1DSD.0D 1D8D.00 1040.00 1020.00 1000.00 DSD.OD

.

Sep bcl yiov but W Veb jta,

ISPX LASTVeekly 03/15/2002 Acme Martel Model(2,20)

Ac leSE Market

-LX - Aclne LEMarlcet

1 >(>(> - (<I/III ll.VO.n

Figure 8.13. S&P 500 hidex U3A)1 - 03/02)

Figure 8.13 IS an extension oftiie chart in Figure 8.12. A bullish "LT" bar occurs in eariy March 2002, and a long entrj is triered the following week with these confirmations:

- High Confirmation for New Lows

- Higli Confirmation for the TRIN

The chart in Figure 8.13 is unfolding as oftiie time ofthis writing. The profit target is not shown, but the stop loss denoted by LX- has been established at an index price of-1140. Further, just after the long position was opened, anotiier LT bar occurred the following week (last bar on the chart), with an LE M Stop justabm.e 1180.

Table 8.9 shows the results for the S&P 500 from 1985 to 2002. 0\er tiiis period, there were 37 weekly signals based on a pattem minimum oftwo and a study length of 20 weeks. O-er a period of800 weeks, that equates to about one signal e-erjfrve months. Eventually, this sjstem wiU be adjusted to a daily time frame when all ofthe data are a-ailable alectronically. Certainly, the sjstem can be run for the subset of indicators that are available on a daily basis.

" w.4-l< ..ll.-. ibis.l,.,,,i<tw.,m.,ii.-...v.i...,.111.,-. I.ll,.........u.-,\i\,, «,vhm(M.«4h in"-

"" I M Si........<. ..... ,.. ,. ,. ...1 ,i- 4.4.1- -.m ,...............,. , ,1,.

1"......v-.l....,l., .,.,., .,, iiM) /\.............,1.......-i........,.........! ..L

1.1II1 ..> ill. w. ,1: ill. .a, .ill..-.»i , ll,,- ., .,ij lwb W\\ I......



Max intraday drawdown ($9,772.00)

Profrt Factor 5.28 Max # contracts hetd

Account size required $9.772 0 Return on account

Table 8.10 shows the results of running the Market System on a 30-week cycle. The profit Factor on the 30 week cycle is 5.28, an improvement over the profit

Table 8.9. Perfbmiance (20) - Acme Man<et Strategy $SPX-Weekly (11/29/1985-3/1/2002)

Total Net Profrt

$94.895 0

Open position P/L

$0

Gross Profit

$158.545.00

Gross Loss

($63,650.00)

Total # oftrades

Percent profitable

54.05%

Number winning trades

Number losing trades

Largest winning trade

$19.926

Largest losing trade

($11.568.00)

Average winning trade

$7.927.25

Average losing trade

($3.744.12)

Ratio avg win/avg loss

2.12

Avg trade (wri & loss)

$2,564.73

Max consec. Winners

Max consec. losers

Avg # bars in winners

Avg # bars in losers

Max intraday drawdown

($20.388 )

Profit Factor

2.49

Max # contracts held

Account size required

$20.388.00

Retum on account

465.45%

Table 8.10. Performance (30) - Acme Market SIrategj- SSPX-Weekly (11/29/1985-3/1/2002)

Total Net Profrt

$112.270

Open position P/L

$0.00

Gross Profrt

$138.51

Gross Loss

($26,240 0)

Total* of trades

Percent profitable

68.18%

Number winning trades

Number losing trades

Largest winning trade

$19.935.00

Largest losing trade

($7,328.00)

Average winning trade

$9,234.00

Average losing trade

($3.74857)

Ratio avg win/avg loss

2.46

Avg trade (win & loss)

$5,103.18

Max consec. Winners

Max consec. losers

Avg # bars in winners

Avg # bars in losers

factor of2.49 on the 20-week cjcle. By lengthening the study cjcle, the niunber oftrades has been reduced from 37 to 22.

Think ofthe market cjcle in terms of each of oiu-sentiment indicators. The trader makes the choice ofhow often each indicator wiU be making new highs and lows based on the cjcle length. As the cjcle is shortened, the number of tradmg signals increases with a decrease inprofit fector. Similarly, as the cjcle is lengthened, the number of signals decreases with anincrease inprofit fector.

If a trader cannot use the S entiment Model on a daily basis, the model is still useflilforidentiingsigrttficaritmarket tuming points. Figure 8.14 shows a subtle example ofa bullish "HLS" bar in June 1998. Typically, a bullish bar wiU occur at the end of a downtrend, but here it occurs after eight weeks of a trading range. From the price bars alone, there are no apparent of an impending raUj; however, thebaris confirming the fohowing:

a Low Confirmation for New Highs

a High Confirmation for New Lows

a High Confirmation for Short Sales Ratio (this is the key)

SSPXLAST4Veekly D8AI4/1S9S

A neLEMarkn

1200.00 1180.00 -1100.00 1140.DQ 1120.00 1100.00

L1O8O.O

Fisurt 8.14. S&P 500 Index June ]

The following week, a long entry triggered, and the market rallied 6.5% over the next month. The key to the trade was the Short Sales Ratio it is an example of

liii.i)-,.....Ill ii.iilii ill ilii-.4 lliiiiiii iii.il.iii.<,i>



[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]