· Analytic
Z32615 LC Hawkes Jump-Cluster Signal
Rolling Hawkes-style jump-cluster intensity on CME Live Cattle — Grade B market-microstructure overlay capturing endogenous volatility clustering regimes.
At a glance
Current reading
-0.581
Calm Regime moderate-negative firing
Signal state
Moderate NEG
phase -1, LC jump-intensity below 2-year baseline
Sample
Daily, 36-year span
LC Hawkes endogeneity, 1990+
Historical IC
-0.059 (p < 1e-7)
Spearman vs LC fwd60d; non-overlap test p=0.055 borderline sig
Baseline LC fwd60d
+1.02%
Unconditional mean
Calm Regime fwd60d
+1.42% mean
+0.40 pp above baseline, 61% hit rate
As of Oct 30, 2026. Updated nightly.
OBJECTIVE
Price jumps in financial markets cluster. The Hawkes self-exciting process captures this: large moves increase the near-term probability of further large moves. The endogeneity parameter — essentially the "branching ratio" — quantifies how much of current activity is self-induced versus exogenously driven. When the Hawkes endogeneity is low, the market is in a quiet regime — price moves are mostly driven by exogenous information rather than self-reinforcing cascades. When endogeneity is high, the market is in a chaotic regime — jumps breed more jumps, flow dynamics dominate information. This signal reads a rolling Hawkes-style jump-cluster intensity on CME Live Cattle, z-scored on 504 trading days, and bounds with tanh. Positive composite = HIGH intensity = chaotic regime. Negative composite = LOW intensity = calm regime. State thresholds at plus or minus 0.2 (moderate) and plus or minus 0.5 (extreme). Daily release cadence. The memorable finding: calm regimes (negative composite) empirically deliver bullish forward LC returns. Chaotic regimes deliver flat-to-slightly-negative forward returns. Daily Spearman IC against LC fwd60d is minus 0.059 at p-value < 1e-7 across 9,049 observations -- significant daily. Non-overlapping (every-60-day) independence test at p-value 0.055 across 150 independent samples -- BORDERLINE significant. Phase +1 (Chaotic Regime) delivers +0.82% fwd60d (-0.20 pp edge below baseline, hit 55%). Phase -1 (Calm Regime) delivers +1.42% fwd60d (+0.40 pp edge above baseline, hit 61%). Phase 0 delivers +0.52%. Grade B market-microstructure overlay. Current reading: Calm Regime moderate-negative firing at composite -0.581. LC jump-cluster intensity has moderated below the 2-year baseline -- quiet market microstructure regime. Historical Phase -1 delivers modestly above baseline. Read on for the historical anchors and how to pair with the LC fundamental stack.
LIVE CHART
LC Hawkes Jump-Cluster Signal
Composite reads -0.581 as of October 30, 2026, in the Calm Regime moderate-negative band near the EXTREME NEG threshold. LC jump-cluster intensity is below the 2-year baseline. CME Live Cattle (orange) has traded in the \$220-230 range.
COMPOSITE
-0.581 / tanh
STATE
Calm Regime
PHASE
-1
FWD 60d
+1.42% (baseline +1.02%)
SAMPLE N
9,049 days
STATE-CONDITIONAL EDGE
Forward 60-day returns by state
FORWARD 60-DAY RETURNS BY STATE
Mean forward-60-day benchmark return per composite state. Baseline (horizontal line) is the unconditional sample mean. Bars above the line are the constructive bands; bars below are the bearish bands.
TODAY’S PLAYBOOK READING
STATE
Moderate Neg
STATE AND HISTORY
Calm Regime moderate-negative firing, 45 days running. Phase -1 historically delivers mean fwd60d LC +1.42%, hit 61%, +0.40 pp above baseline. Phase +1 (Chaotic) delivers +0.82% / 55% hit / -0.20 pp below baseline.
THE MICROSTRUCTURE MECHANISM
Low Hawkes endogeneity = market in quiet regime (moves driven by information not cascades) = constructive-flow forward LC. High endogeneity = chaotic regime (jumps breed jumps) = flow dominates, forward returns flatten.
HOW TO POSITION
Pair with LC-corn ratio, FC-LC spread, cattle inventory, term structure. Hawkes jump-cluster is a market-microstructure overlay distinct from supply/demand fundamentals — captures flow-regime context.
WHAT THE CHART SHOWS
The composite sits at -0.581 near the Calm Regime EXTREME NEG threshold. The blue signal line has moved below -0.5 as LC jump-cluster intensity moderated through October. CME Live Cattle (orange) has traded in the \$220-230 range.
TOP INPUTS FIRING
- Primary composite -0.581 State Calm Regime moderate-negative firing, 45 days running.
- LC jump-cluster intensity moderated Rolling Hawkes-style endogeneity on LC daily returns is below the 2-year baseline — quiet microstructure regime.
- State-conditional edge: +0.40 pp mean Calm Regime historical fwd60d LC mean +1.42%, baseline +1.02%, hit 61%. Current firing near EXTREME NEG threshold — modestly constructive microstructure read.
THE TAKEAWAY
What this means for you today
THE MICROSTRUCTURE CALL
Composite -0.581, Calm Regime moderate-negative firing. LC jump-cluster intensity has moderated below the 2-year baseline — the market is in a quiet microstructure regime.
THE EMPIRICAL CONTEXT
Calm Regime historically delivers mean fwd60d LC +1.42% (+0.40 pp vs baseline, hit 61%). Current firing is on the EXTREME NEG threshold (-0.5) — stronger signal than Moderate.
THE HONEST CAVEAT
Daily IC -0.059 significant. Non-overlap -0.157 at p=0.055 BORDERLINE significant. Grade B microstructure overlay. The signal reads flow-regime context, not direction.
THE REGIME MAP
The three states, forward returns and sample share
Computed from 9,049 trading days with observable LC fwd60d. Baseline LC fwd60d (unconditional): plus 1.02 percent mean. Grade B market-microstructure overlay with calm-regime positive edge.
| State | Phase | Share | Fwd 60d | HAC t | Edge vs baseline | Last firings |
|---|---|---|---|---|---|---|
| Chaotic Regime | +1 | 34.9% (3159 d) | +1.01% | -0.01pp | Sep 2026, Aug 2026, Aug 2026 | |
| Normal TODAY | 0 | 20.7% (1876 d) | +0.72% | -0.30pp | Oct 2026, Sep 2026, Aug 2026 | |
| Calm Regime | -1 | 44.4% (4014 d) | +0.97% | -0.06pp | Oct 2026, Aug 2026, Aug 2026 |
The historical pattern
Signature episodes in the 36-year sample
What this measures
The signal computes a rolling Hawkes-style jump-cluster intensity on CME Live Cattle daily returns (Truth LC_T050, n*=0.636 target endogeneity). Z-score on 504-day rolling window with 60-day minimum sample. Tanh-bounded to [-1, +1]. State thresholds plus or minus 0.2 (moderate) and plus or minus 0.5 (extreme). Daily release.
Why it matters
Why Hawkes endogeneity reads market microstructure
The Hawkes self-exciting process is a well-studied model of event clustering. Applied to price jumps, it decomposes observed volatility into an exogenous component (driven by news and fundamentals) and an endogenous component (jumps breeding further jumps through self-reinforcing cascades). The branching ratio n* measures the share of activity that is self-induced. High n* means the market is in a flow-dominated regime where news effects are amplified by cascade dynamics; low n* means orderly information-driven price discovery.
Why Grade B borderline-independent
Daily Spearman IC against LC fwd60d is -0.059 at p-value < 1e-7 across 9,049 observations -- significant daily. Non-overlapping (every-60d) independence test at p-value 0.055 across 150 independent samples -- BORDERLINE significant. Phase -1 (Calm) delivers +0.40 pp above baseline, Phase +1 (Chaotic) delivers -0.20 pp below. Grade B market-microstructure overlay distinct from all supply/demand fundamentals.
How to use this signal today
Pair with LC-corn ratio, FC-LC spread, cattle inventory, term structure, slaughter YoY. Hawkes jump-cluster is a market-microstructure overlay that captures flow-regime context — distinct from all supply/demand fundamentals. When the market is in calm regime, fundamental signals have cleaner transmission to forward returns.
Related signals
Not investment advice · The analysis on this page is informational and educational. Signal readings, historical case figures, and empirical statistics are derived from publicly available data and are provided without warranty. Hypothetical performance results have inherent limitations, including the benefit of hindsight. Readers who intend to trade should consult a licensed investment professional and see the research brief for the full disclaimer.
Last refreshed Nov 01, 2026 01:09 PM ET