The platform I’ve been using and building on since NT6.5. Setup, troubleshooting, and what makes it work.
A Kalman filter is the natural response to noisy price data — recursively estimate the underlying state and downweight new observations by how much you trust them. BigBeluga’s Kalman Volume Trend takes that idea, applies it to close, and stacks volume context around the resulting trend line: per-bar volume delta bars sized by the normalized…
“How many bars since the last cross?” “How many bars has price been above the SMA?” Bars since a condition was met is one of the most-reused small patterns in NinjaScript — a single int field that increments every bar and resets to zero whenever the trigger fires. This post covers the pattern, the corner…
The previous post built a Random Forest — many decision trees, each grown independently on its own slice of the data, their votes averaged. This post keeps the trees but changes how they are built and combined. Gradient boosting grows trees in sequence: each new tree is fitted to the errors the ensemble still makes,…
SuperTrend gives you a binary trend signal: above the band or below it. ChartPrime’s SuperTrend Oscillator turns the same primitive into a graded momentum read by measuring the distance from price to the SuperTrend line, smoothing it, and normalizing — turning a yes/no flip into a continuous oscillator that swings roughly in ±1.7. The result…
Running highs and lows are the building blocks of breakout indicators, pivot detection, stop placement, and risk sizing. “Highest high of the last 20 bars” is one function call. “Highest high of this session so far” is a loop. “Highest high since the last SMA cross” is a reset-on-trigger pattern. Different problems, different tools —…
Posts 2 through 7 of this series were one long thread. A single neuron became hidden layers, hidden layers became configurable depth, depth got a better optimizer, and then the network grew a memory and finally gated memory. Every one of those models learned the same way: compute a gradient, nudge the weights. This post…
Norman Fosback’s Positive Volume Index (PVI) and Negative Volume Index (NVI) are old-school: PVI accumulates only on rising-volume bars (the crowd, the dumb money), NVI only on falling-volume bars (the operators, the smart money). The classic read is that NVI rising while PVI is flat = quiet accumulation. AlgoAlpha’s Smart Money Volume Index takes that…
You have an SMA. You want to know when price crosses it — that’s a signal. You also want to know when the SMA itself changes direction — that’s a different signal, same class of problem. Both come down to comparing current and previous values, and NinjaScript has built-in helpers for the common case plus…
The previous post built a vanilla Recurrent Neural Network — a model with a hidden state that carries forward bar to bar — and then ran straight into its defining limitation. Training a recurrent network means multiplying the gradient by a tanh derivative at every step back through time, and a long chain of numbers…
Standard SuperTrend gives you a clean trailing stop in trends and brutal whipsaws in noise. LuxAlgo’s Evasive SuperTrend is built for the noise problem — when price gets within a configurable Threshold × ATR of the active band, the band actively pushes away from price by Alpha × ATR, dodging the wick that would have…
Notifications
