Constructing ADX, RSI, and MACD price filters

A three-layer construction drill

This article is a construction drill for three price filters. The archive defines the Average Directional Movement Index as a measure of market trend intensity, the Relative Strength Index as a tool for identifying overbought or oversold conditions and divergences, and MACD as two exponentially smoothed moving averages plotted above and below a zero line.

Editorial interpretation: finish each object as a rule that later observations can accept or reject, rather than as a chart decoration.

Trend intensity from directional movement

The Average Directional Movement Index is defined as a measure of market trend intensity. The matching construction is a quantitative reading of that intensity built from directional price movement rather than from price level alone.

Editorial interpretation: use the finished reading as a trend-intensity filter that can stand in front of other conditions.

A bounded overbought-oversold oscillator

The Relative Strength Index is defined as a tool for identifying overbought or oversold conditions and divergences. The matching construction is a bounded oscillator used to mark those conditions and to compare price swings with indicator swings for divergence.

Editorial interpretation: keep this layer bounded so overbought and oversold states remain explicit conditions, not informal chart comments.

A zero-centered spread of two averages

A moving average is defined as a mathematical smoother that reduces fluctuations, emphasizes trend direction, and can confirm reversals. In the same construction language, it replaces successive observations with an average of recent values to emphasize direction and reduce short-term noise.

MACD is defined as two exponentially smoothed moving averages plotted above and below a zero line. The matching construction is a zero-centered comparison of those two averages.

MACD signal conditions are defined as average-line crossovers, movement through the zero line, and divergences. A moving-average crossover is defined as the intersection of average lines with each other or with price and is treated as a price-based opportunity condition.

Finished objects as falsifiable rules

Backtesting is defined as testing or optimizing a strategy on historical data and then applying it to new data to check consistency. The same workflow evaluates or fits a rule on historical observations and then applies that rule to later observations to check whether results stay consistent.

Editorial interpretation: after the three layers are assembled, write each filter as a statement about trend intensity, bounded conditions, or average-spread behavior, then apply the same statement to later observations. The point of the drill is a falsifiable rule, not a decorated chart.