Mnikolic Investment Technologies
intelligent investment technologies™
Mnikolic Investment Technologies
intelligent investment technologies™

PivotBuilder Personal Trading Suite v1.10 introduces a major new feature aimed at weighing pivot slope based on regression analysis.

INTRODUCING LINEAR REGRESSION

In financial trading, linear regression is often considered a major trend line, albeit one based on price distribution (pure statistics) rather than chart patterns. You would treat a regressive trend line as you would a pattern one - price above it is bullish, below bearish, with crossovers signifying a change in sentiment. Unlike pattern trend lines, statistically-derived trend lines like those utilizing linear regression, can also outline overbought/oversold market conditions if current price is too distant from them. If price is extremely above (relative to history) regression-derived trend line, it is considered overbought. If it is extremely below, it is considered to be oversold. The graph below is intended for those new to the concept of linear regression. You can think of linear regression as a central mean line representing a center of price "mass" for the past x amount of bars. Price far from its linear regression slope is expected to retrace.

LINEAR REGRESSION IN PIVOTBUILDER v1.10

PivotBuilder v1.10 takes the concept I described in the prior paragraph and incorporates it into pivot calculation, making it more adaptive and applicable to trading. In this version of PivotBuilder, Line1 pivot (assumed to be the central, non-deviated pivot) has an extra parameter that allows the user to apply linear regression into its calculation. That parameter is called Line1.Regression and has a value input range of 0. Instead of overriding PivotBuilder's adaptive algorithm with a full-fledged linearly regressive one, we opted to instead give user the option to choose how much linear regression should be weighed during pivot calculations. A value of 0 does not consider any linear regression in pivot calculation. The higher the value, the greater the weight given to linear regression, with 1.0 representing a near-full regressive behavior within Line1 pivot calculations. We suggest you start with 0.1 and go on up from there. Below are 2 comparative screenshots to help you see how higher Line1.regression value impacts the Line1 (central) slope. Note: only a single pivot line was activated and drawn for simplicity and demonstration purposes. As you know, PivotBuilder can draw up to 7 independent pivot slopes (see Number.of.pivot.lines parameter).

E-MINI S&P 500 (CME:ES) 15-MINUTE, LINE1 LINEAR REGRESSION SET TO 0.25 (purple slope line)

E-MINI S&P 500 (CME:ES) 15-MINUTE, LINE1 LINEAR REGRESSION SET TO 0.80 (purple slope line)

HIGHER PERIOD = HIGHER REGRESSION

In addition to the Line1.Regression weighing parameter mentioned in the prior paragraph, the Line1.Period parameter, representing historical length of Line1 pivot, plays an equally important role when applying linear regression. The higher the period, the longer the historical length considered when calculating the Line1 pivot. Higher period values typically translate to smoother, and consequently, less responsive Line1 pivot (the distributed set is vertically wider).

EXTENSION IN v1.11

PivotBuilder v1.11, scheduled for early 2014 release will extend linear regression weighing to other 6 pivots.

RELEASE NOTES AND TUTORIAL

Full details pertaining to PivotBuilder Personal Trading Suite v1.10 release, including complete feature list, and a tutorial can be accessed via official release post.

LEARN MORE ABOUT V1.10

EXISTING CLIENTS

  • Version 1.10 framework, like prior upgrades, is free to all existing PivotBuilder license holders.
DOWNLOAD VERSION V1.10

NEW CUSTOMERS

  • If you are interested in purchasing PivotBuilder, please visit the official store page below.
PURCHASE PIVOTBUILDER

Back to top