MarketScalpel's cycle analysis tools are designed to detect whether periodic cycles are present in sector price changes, and if so to estimate their duration, and the time remaining until the next high and low in price are expected. This data in effect provides tactical position management support, assisting managers and traders in maximizing performance by identifying optimal times to establish, exit, add, or trim positions.
We use cycle analysis for sector forecasting in conjunction with various proprietary sector internals data and analysis, including our Volume Confirmation signals. ; For further theoretical background information on the subject of cycles please see a basic introduction to the principals of cycle analysis and terminology.
The Market Navigator research platform uses a spectral analysis algorithm with a fractal filter to identify frequency response and, by extension, the presence of cyclicality in our sector price indices. This technique is derived from mathematical analysis originally developed for digital signal processing to filter noisy frequency data.
The analysis seeks to identify cycles (and associated frequencies) that provide the greatest amount of regular, periodic price movement at the sector level. Our spectral analysis algorithm has the advantage of working well with small data samples making it highly responsive to changes in the dominant cyclicality.
The Market Navigator sector charts provide two dominant cycle estimates; a long cycle and a short cycle. These are graphically represented in the default sector chart (see A3 in the controls overview) by oscillators with the long cycle colored green and the shorter cycle orange.
Additional text-based information necessary for evaluation of the cycle estimates is presented in the legend to the right of the chart, as well in the Cycles related Sector Overview table.
The ; Days To Top estimate is supplied as a rough guide that should be treated with some caution. ; Cycles often exhibit left or right translation meaning that the top occurs late or early relative to the cycle midpoint in time. ; This effect is often present even where a cycle is well-defined in terms of the bottom-to-bottom symmetry, which is the more stable measure of cyclicality.
It is instead strongly recommended to monitor for cyclical highs via: