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]
52 the same time, then the system in question may not be as good as first thought. Under the same circumstances, if the system trades less than 100 percent of the time, then its performance is subject to interpretation. Less exposure to trading translates into less risk associated with trading the system. The importance of time evaluation is discussed later in the chapter. Performance should not be valued based on a single return measure. All these return measures play a role in the evaluation process. The importance of each measure depends on the system and its style of trade. Sliding and Rolling Summaries This section expands on the general overview of a systems trading performance. In the previous sections, the evaluation tools measured performance of the test period from start to end. The next step is to examine the system over various time periods to assure consistent performance during the trading process. After all, what good is a net profitable trading system if a trader goes broke after an early major loss? Remember that consistency breeds confidence. Trading summaries can be based on any time frame. However, this report will examine monthly and annual periods. This detailed evaluation flushes out hidden strengths and weaknesses not readily apparent in a general overview. A trading summary can consist of the following evaluation methods: • Sliding Analysis. Searches for pockets of strength or weakness by examining short periods of time, one after the other. • Rolling Analysis. Searches for sensitivity to starting dates by examining trading periods with the same end date but with different starting dates. • Equity Curves. A graphical illustration of the systems profit/loss performance over time. The example of two trading systems will help illustrate the advantages of examining a system over various time frames. These examples will focus on annual data; if greater detail is required, we suggest a monthly breakdown. Intraday traders should use daily and hourly breakdown periods to assure a thorough evaluation. Both systems in Table 9.4 generated the same net profit over the same time period; however, their paths were distinctly different. Based on annual sliding analysis, System As profit taking was consistent and System was volatile. Based on rolling analysis, if System was tested only on data from 1993 through 1996, then its overall net profit of $600,000 would be three times as great as that of System A. Nonetheless, we know better than to put our money on B. After all, the largest rolling period shows both systems ultimately delivered the same net profit in a 7-year span. It is best to avoid systems that generate volatile trading results. At a minimum, traders should be acutely aware of past volatility and, therefore, better prepared to trade the system into the future.
| Annual Sliding Analysis | | | Annual Rolling Analysis | | | System A | System | | System A | System | 1990 | | -50K | 90-96 | 350K | 350K | 1991 | | -100K | 91-96 | 300K | 400K | 1992 | | -100K | 92-96 | 250K | 500K | 1993 | | 350K | 93-96 | 200K | 600K | 1994 | | -50K | 94-96 | 150K | 250K | 1995 | | 275K | 95-96 | 100K | 300K | 1996 | | | | | | Net profit | 350K | 350K | Net profit 350K | 350K |
Both sliding and rolling analysis could process other evaluation measurements, such as percent gain, profit factor, and percent profitable. Equity Curve Analysis Viewing a systems equity curve can provide additional insight into its performance (see Box 3). Equity curve charts tally a systems net equity on a trade-by-trade basis. Until a trader sees a systems equity curve, he or she will not have a big picture of system performance. A quick review of an equity curve chart can provide the necessary mental security to trade a system. Equity charts come in a variety of formats. Each chart centers on the same basic information but presents the data in a different manner. Since traders often BOX 3 . EQUITY CURVES Equity curve line: Plots the closed realized net profit per trade without regard j to time. Detailed equity: Plots the mark-to-market net profit result per bar including ! flat or nontrading periods. Monthly rolling equity: Plots the mark-to-market net profit result per month. j Average monthly equity: Plots the average net profit by month. : Underwater equity: Plots a monthly equity curve with an emphasis on the magnitude and duration of percentage drawdowns between equity peaks. TABLE 9.4 TRADING SUMMARY COMPARISON
have different trading expectations, it is best to evaluate trading performance with multiple equity charts. Each of these five equity charts illustrates a different aspect of trading. Reviewing all these charts paints an evaluation picture not readily seen in other formats. Figures 9.1-9.5 show five of the more important equity curve charts. Equity Curve Line The equity curve line chart in Figure 9.1 presents trading performance on a trade-by-trade basis. This chart does not use time on its X-axis, but rather the trade number (1st trade, 2nd trade, etc.). This all-purpose equity chart is best used as a general snapshot of trading performance. Detailed Equity Curve The detailed equity curve in Figure 9.2 offers greater insight into trading performance than a general equity curve chart. This chart displays net profit on a bar-by-bar basis revealing full equity drawdowns and run-ups. Flat or nontrading periods are also shown to present a detailed overview of equity performance. Rolling Period Equity Curve The rolling period equity curve calculates realized or unrealized profits or losses at specific periodic times. For example, an end-of-month analysis is shown in Figure 9.3. Figure 9.1 General equity curve. RINAl qNSP 460000-Daily (08/19/85-07/18/97) 600000 500000 - g 400000 - J I300000- jr 200000 - n -- ---- 100000 - vc-o-0-""" 0 l---,-,-,-,-,-,-,-1 1 11 21 31 41 51 61 71 81 91 number of trades
[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]
|