Pattern Matcher

Pattern recognition and similarity search for forecasting.

Find Similar Patterns

matcher = SQA::PatternMatcher.new(stock: stock)

# Find historical patterns similar to current
similar = matcher.find_similar(
  lookback: 10,           # 10-period pattern
  num_matches: 5,          # Find 5 best matches
  method: :euclidean      # Distance metric
)

similar.each do |match|
  puts "Distance: #{match[:distance]}"
  puts "Future return: #{match[:future_return]}%"
end

Distance Methods

• :euclidean - Standard distance
• :dtw - Dynamic Time Warping (handles compression/stretching)
• :correlation - Correlation-based similarity

Forecasting

forecast = matcher.forecast(
  lookback: 10,
  forecast_periods: 5,
  method: :dtw
)

puts "Forecast price: #{forecast[:forecast_price]}"
puts "Expected return: #{forecast[:forecast_return]}%"
puts "95% CI: #{forecast[:confidence_interval_95]}"

Chart Pattern Detection

# Detect double top/bottom
patterns = matcher.detect_chart_pattern(:double_top)

# Detect head & shoulders
patterns = matcher.detect_chart_pattern(:head_and_shoulders)

# Detect triangles
patterns = matcher.detect_chart_pattern(:triangle)

patterns.each do |pattern|
  puts "#{pattern[:type]} at index #{pattern[:index]}"
  puts "Confidence: #{pattern[:confidence]}"
end

Pattern Clustering

# Group similar patterns
clusters = matcher.cluster_patterns(
  num_clusters: 5,
  lookback: 10
)

clusters.each_with_index do |cluster, i|
  puts "Cluster #{i}: #{cluster.size} patterns"
  puts "  Avg future return: #{cluster[:avg_return]}%"
end