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] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [ 117 ] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] [178] [179] [180] [181] [182] [183] [184] [185] [186] [187] [188] [189] [190] [191] [192] [193] [194] [195] [196] [197] [198] [199] [200] [201] [202] [203] [204] [205]


117

If - Wed[1xl then n3[i><] - n3[<x] ♦ 1; end;

for ix = 1 to 8 begin

if day4 - fhursCix] then n4[ix] - n4[ix] * 1; end;

for ix - 1 to 16 begin

if day5 - Fri[ix] then oStixl - n5[ix] * 1; end;

end;

end;

If lastcalcdate = date ther begin

prirt ("Monday = 1 100 . cases nion;5;C); prirt ("Tuesday); for ix - 1 to 2 begir

div = @irtportiont(ix+l)/2);

PC - 211 100/ ;

print (Tuestix]:5:0, n2tix]:4;0, pc;4;0); end;

print ("Wednesday"): for ix = 1 to 4 begin

div - eintportior((ix+l)/2);

pc - n3[ix]*100/n2Idiv];

print (Wed[ixl;6;0, n3[ix]:4:0, pc:];0);

end;

print ("Thursday"!; for ix - I to e begin

d1 V = ©intporti on( M"x+1 1/ 1; pc = n4[ix]*100/n3[div];

print fThurs[ix]:5:0. n4[1x]:4:0, pc:4:0); end;

print ("Friday");

for TK - 1 to 16 begin

div = @intportion(fix-l)/Z);

pc = n5[ix]*10D/n4[div];

print (Fr1[1x]:5:D, n5[ix]:4:0, pc:4:D);

end;

end;

Weekend Pattems

Of all pattems studied, this test was expected to be the most promising. Two factors lead to the anticipation of sttong weekend pattems:

1. Friday liquidation-the aversion of tiudeis to hold a market position over the weekend.



2. Setting of new positions and resetting of old ones on Monday, especially during a trending market.

if the pattems can be identified, the 2-day weekend delay should result in large price jimips and good profits for those willing to take the weekend risk. Not knowing what to expect but anticipating combining the results with the study of weekday pattems, a ramiber of different trend indicators and pattems were tested These were limited to closing price relationships.

Trend indicators

[*rior Thursday-to-Friday direction

Friday-to-Friday direction

Most fi-equent direction last week

Following each of the tiends or pattems, the prices on Monday were tabulated to determine the fi-equencj and consistencj of the opening price, closing price, and high or low prices that were in the same direction as each of the tiends or pattems. Table 15-7 lists the results for the six markets studied for the 10 years fi-om 1986 through Februarj 1996. Based on the three tiend pattems, none of the markets show any significant tendencj to carrj the tiend over the weekend reflected in the open or close of the following Monday. The extreme on Monday, that is, the high or low pnce, normally provides an opportunity to exit at a profit fi-om a tiend position taken on Fridajs close sometime on Monday.

ft seems surprising that the price pattem on Monday no longer reflects the direction of the pre™us Thursdajto-Friday move, nor the tiend of the pre™us week. In the same study performed in the prior 10 years, there was a clear pattem based on Thursday to Friday. Considering the fact that there have been dominant tiends in the equities markets and interest rates, the only reasonable conclusion is that markets are more complex, based on greater participation, it will be necessarj to look at longer tiends or other pattems to find high-probability cases.

Combining Weekend and Weekday Pattems

The direction of Mondajs prices is more interesting when viewed in terms of the pre\ious weekday pattems. Seen side by side, the fi-equencj of each pattem and the way the six sample maikets react with respect to the following Monday closing price give insight into mar

TABLEl 5-7 Weekend Pattem Based on Possible Trends during tiie Pnor Week, Januarj 1986-Februarj 1996



Continued Direcuon Monday (%)

Market

Trend

Open

Ctose

Extreme

Deutschemark

Thurs to Fri

Fri to Fri

Most frequent direction

S&P500

Thurs to Fri

Fri Fri

Most frequent directior

U.S.T-bonds

Thurs Fri

Fri Fri

Most frequent direction

Heating oil

Thurs Fri

Fri 00 Fri

Most frequent direction

Gold

Thurs Fri

Fri Fri

Most frequent direction

Soybeans

Thurs Fri

Fri Fri

Most frequent direction

ket behavior Table 15-8 shows the percentage of total cases for each weekday pattern (based on about 480 weeks) and the percentage of days for which Mondays price closed in the X direction, continuing the direction of the previous Monday.

In Table 15-8 there are two mteresting items to observe, the frequency of pattems and the percentage of cases that Monday follows specific pattems. The most consistent pattem seems to be the low percentage associated with XXXXX, the case in which the market moves the same direction every day. While it appears most often in the Deutschemark, considered the most trending of these markets, it is not a common pattem in the other markets. When it does occur, the D-mark and heating oil are most hkely to reverse direction on Monday, while the U.S. Treasury bonds and soybeans are hkely to continue the trend.

Overall, the fiuctuating pattems are more common than persistent direction. For example, XOXOX and XOOXO are among the most common pattems. These both show Tuesday reversals and the results can be compared with the weekday study. While there is consistency in the reaction of the financial markets, soybeans tend to move in the opposite direction.

Specific market pattems, such as weekday and weekend, can provide an edge for traders in a very different and more complex market enonment. By finding pattems that are more hkely. or even less hkely, to occur over groups of similar markets, with 10**o greater fiequencj, traders can - their position size based on expectations and thereby gain a market advantage

Programming the Weekend Study

The following Omega Easy Language program prints the results used in the weekend study to the Print Log. Some of the techniques used are the same as those used in the weekday study

TABLE 15-8 Combining Weekend and Weekday Pattems, January 1986-Februai3 1996*

D-marb S&P 500 U. 5. T-Bonds Heating Oil Gold Soybeans

Pattem "oTotal <»oX<»oTotal M "oTotal M "oTotal M WTotal M»6TotalM

XOXXX 6.3 60 6.3 57 6.5 35 57 37 6.8 537.5 42

XOXXO 8.6 44 6.5 48 4.9 39 4.0 32 5.9 so 5.9 32



[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] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [ 117 ] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] [178] [179] [180] [181] [182] [183] [184] [185] [186] [187] [188] [189] [190] [191] [192] [193] [194] [195] [196] [197] [198] [199] [200] [201] [202] [203] [204] [205]