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]


56

weakness is indicated if the market cannot trade higher into the close. Likewise a weak opening followed by further weakness is a continuing negative, while a weak opening followed by strength is positive.

Candlestick charting is hundreds of years old, and the Japanese have learned to focus heavily on the patterns formed by the candles, especially certain groupings of candles. We find it interesting to use the candles themselves to add some extra visual value to Western-style charts by elucidating the price action as only candles can do-especially price action near the Bollinger Bands. For example, a transition from red bodies-where the close is lower than the open-to green bodies-where the close is higher than the open-near the lower band is an indication of a bottom. That transition would be mirrored in Accumulation Distribution, especially if volume rose with the turn in the tide.

Both Intraday Intensity and Accumulation Distribution can be presented in open-ended form by keeping a cumulative sum from the first calculation point of the individual-period readings and plotting the result as a line. They also can be presented in the form



Chapter 18: The Squeeze 151

Table 18.4 Formula for Normalizing Volume Oscillators

10-day sum of [(close - open)/(high - low) * volume]/

10-day sum of volume

of an oscillator by keeping an n-period moving sum of the periodic readings, where is typically 10 or 20 periods. Some people find it easier to read the unbounded indicator line; some people prefer the oscillator format. For comparison to trading bands, the oscillator format seems easier to interpret for most.

Either an II or AD oscillator can be normalized so that it is directly comparable from issue to issue by dividing it by the total volume over the calculation period. Table 18.4 presents the formula for a 10-day normalized Accumulation Distribution oscillator, and Figure 18.3 presents the result. The normalized versions are often referred to as percents, i.e., 21-day 11% or 10-day AD%.

5/00 8/00 7/00 8/00 9/00 10/00 11/00 12/00 1/01 2/01

Figure 18.3 11% and AD%, Hartford Insurance, 200 days. Converting II and AD to oscillators often uncovers data that is hard to see otherwise.



Part IV: Bollinger Bands with Indicators

In the fourth category we find the Money Flow Index, which elegantly imbues Welles Wilders Relative Strength Index concept with volume data. At the core of the RSI calculation is a ratio of two exponential averages, one of the changes on up periods and another of the changes on down periods. MFI replaces those exponential averages with moving sums of price times volume, one of positive days and one of negative days. Common sum lengths are 9 for short-term purposes and 14 as recommended by Wilder in his original presentation.

With MFI you are asking the question, does the volume differential between up days and down days confirm the momentum of the trend? A rally with strong volume on the up days and contracting volume on the down days will produce a stronger MFI than would an RSI of the same period.

MFI turns out to be a bit more volatile than RSI. It does not trend as smoothly, and it exhibits a greater range. Therefore we use the levels of 80 and 20 as MFI benchmarks rather than the standard benchmark levels of 70 and 30 used for RSI. This results in roughly equivalent signals (Figure 18.4).

1 i J

S 4, If

ijjfl \

1 1

1 1

Ay"

5/00 6/00 7/00 8/O0 9/00 10/00 11/00 12/00 1/01 2/01

Figure 18.4 Money Flow Index, Hartford Insurance, 200 days. Note the steadily deteriorating trend in MFI as the rally matures.



[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]