1986issue C081-3
Evaluating advance-decline plus-day runs against chance baselines
Exchange advance-decline plus-minus sequences can be compared with chance-game outcomes and a computer-generated plus-minus path. A runs-test counts plus-day runs so a streak profile can be located against those baselines.
- Exchange advance-decline plus-minus sequences can be compared with chance-game outcomes on a probability basis.
- A runs-test counts plus-day runs in lengths of one, two, three, or more so isolated plus days and longer streaks can be placed beside chance-game counts.
- A Monte Carlo path of computer-generated plus and minus days supplies a random baseline for the same sampling interval, though it was said to miss the emotional persistence that can extend a move.
- Later samples can wander above and below a probability-line drawn from a large trial count; the comparison was not presented as a projection tool.
Why compare breadth with chance
Market-breadth, in this evaluation, is a daily advance-versus-decline tally treated as an ordered plus-minus series rather than as a price-chart pattern. Exchange advance-decline plus-minus sequences can be compared with chance-game outcomes on a probability basis.
A plus-day-run is a streak of consecutive sessions in which advances outnumber declines. The historical workflow used that ordered series as the object of a runs-test, not as a chart pattern to be read by eye.
How the runs-test is counted
A runs-test is a count of consecutive plus or minus observations used to compare an ordered series with a chance expectation over a fixed lookback. Consecutive plus observations were counted in run lengths of one, two, three, or more.
That tally keeps isolated plus days and longer plus-day runs in the same table, so each length can be set beside the matching chance-game count.
Chance-game and Monte Carlo baselines
The tabulated comparison used 6500 trials, and later samples can differ while still wandering above and below a probability reference line. A probability-line is the reference frequency implied by a large trial count, against which later samples can wander above or below.
Monte Carlo simulation, in this setting, means computer-generated plus and minus sequences used as a random baseline for the same sampling interval as the observed series. Computer-generated plus and minus days used as a market simulation were said to miss the emotional persistence that can extend a move once it is underway.
Where the run lengths sat
Isolated plus days in the advance-decline series appeared out of line with the corresponding chance-game counts. Advance-decline runs of six or more consecutive pluses appeared to outpace the two chance-game series.
A later stretch of small fluctuations after a declining market was offered as a possible setting in which isolated plus-day counts might realign. Editorial: that remark is a historical hypothesis about when a short-run count might move back toward the chance-game table, not a rule for acting on the next plus day.
Reading the meander, not a projection
The plotted series meandered above and below the 6500-day result, and the comparison was not presented as a projection tool. Additional figure panels accompany the run-length and meander comparison.
Editorial: a current streak profile is located when its run lengths and its path relative to the probability-line are known. That location is the evaluation step. It is not, by itself, a forecast that the next sessions will continue the same plus-day-run.
All readings on this track · 15 readings
- 1986Constructing runs and persistence tests from labeled prices
- 1986Evaluating daily price and volume serial independence windows
- 1986Evaluating advance-decline plus-day runs against chance baselines
- 1986Weekly resamples as a diagnostic filter for statistical windows
- 1988Runs test as a critique of price-series memory
- 1989Evaluating weekday close direction with a counted baseline
- 1989Statistical windows for indicator time parameters
- 1992Channel-height ratios for equity trend evaluation
- 2001A runs test before volatility and expected-value sizing
- 2005Constructing runs-test z-scores for signed return persistence
- 2005Evaluating persistence with runs and autocorrelation
- 2005Weekday FX turning points and close run tests
- 2013Constructing a runs-test turn forecast
- 2017Star rating from slope and swing runs
- 2018Regime-dependent odds after directional price runs