GC · Metals / Cross-Metal Variance Spillover
Z13068 – GC Gold Platinum-to-Gold Variance Spillover
Platinum volatility leads gold volatility when PGM-complex stress transmits across metals; the 1-day-lagged variance spillover coefficient captures cross-metal transmission on a 60-day forward horizon.
At a glance
IN-SEASON · ACTIVE
Current reading
-0.120
1-day-lagged variance spillover PL to GC, tanh-bounded
Signal state
Neutral
167 trading days in current state
Phase
NEUTRAL
Positive equals elevated spillover; negative equals calm
Sample
16 years
Since 2010, roughly 4,231 trading days
As of Oct 30, 2026. Updated nightly.
SIGNAL OVERVIEW
Platinum and gold both trade in the precious-metals complex but the two respond to different fundamentals. Platinum carries autocatalyst demand exposure; gold carries safe-haven and monetary exposure. When cross-metal stress episodes hit — flow-driven liquidity shocks, speculative unwinds, PGM-specific supply shocks — platinum volatility leads gold volatility with a 1-day lag. This signal captures the asymmetric transmission. The engine computes a 1-day-lagged variance spillover coefficient from platinum squared returns to gold returns, rolling window regression, and bounds through tanh. On March 19 2020 the composite reached series maximum plus 0.9036 as COVID liquidity-shock variance spilled from platinum to gold. On July 26 2023 the composite reached series minimum minus 0.5677 in a calm cross-metal regime. Today the composite reads -0.120, in Neutral territory.
LIVE CHART
The 504-day z-score of a multi-input composite of US credit-quality indicators including BAA-AAA corporate spreads, 9016 trading days of history for GC.
The dark line is the composite bounded between minus one and plus one. Pale red marks spillover-stress; pale green marks spillover-calm. Bronze line is the trailing 252-day rolling Spearman IC. Lower panel is COMEX GC continuous back-adjusted price on the right axis.
COMPOSITE
+0.804 / tanh
STATE
Extreme Pos
PHASE
+1
FWD 60d
+5.24% (baseline +2.32%)
SAMPLE N
9,016 days
TODAY'S PLAYBOOK READING
STATE
NEUTRAL
STATE AND READING
Neutral, 167 days running. Composite at -0.120 is in the range.
WHY THE SIGNAL IS HERE
The underlying input is at its current level relative to the 504-day baseline. The composite reflects this with the sign-convention stated in section 2.
POSTURE FOR THE WEEKS AHEAD
The signal's historical edge is concentrated at the tanh-saturation extremes (composite below minus 0.5 or above plus 0.5). The cross-sample Pearson IC of +0.11 on SPY forward-sixty-day is a reference-asset read; the appropriate direct read is GC forward returns, which are not computed as a baseline in this page.
WHAT THE CHART SHOWS
Today the composite reads -0.120 placing the dot inside the Neutral band. The spillover coefficient updates daily from the rolling window regression; watch for sustained elevation as cross-metal stress signal.
TOP INPUTS FIRING
- Platinum squared returns Lagged 1 day; the volatility-source variable in the spillover regression.
- Gold returns Dependent variable in the spillover regression.
- Rolling regression coefficient Variance transmission from PL to GC with 1-day lag.
THE REGIME MAP
The six regime states
| State | Phase | Share | Fwd 60d | HAC t | Edge vs baseline | Last firings |
|---|---|---|---|---|---|---|
| Spillover stress | +1 | 14.7% (621 d) | +2.34% | +1.04pp | Dec 2025, Oct 2025, Oct 2025 | |
| Normal TODAY | 0 | 76.9% (3254 d) | +1.30% | +0.01pp | Mar 2026, Nov 2025, Oct 2025 | |
| Spillover calm | -1 | 8.4% (356 d) | -0.57% | -1.87pp | May 2025, Jan 2024, Jan 2024 |
The historical pattern
Signature episodes
THE TRACK RECORD
Every state change in the last 2 years
8 state transitions from 01/02/2025 to 11/04/2026. Full history (99 total transitions since 11/30/1990) is downloadable.
| State | Start | End | Duration | Fwd 60d GC |
|---|---|---|---|---|
| Neutral | 01/02/2025 | 01/31/2025 | 21d | -4.32% |
| Defensive | 02/03/2025 | 02/28/2025 | 19d | -7.23% |
| Neutral | 03/03/2025 | 07/31/2025 | 105d | +0.68% |
| Defensive | 08/01/2025 | 08/29/2025 | 21d | +10.22% |
| Neutral | 09/02/2025 | 10/31/2025 | 44d | +5.43% |
| Expansion | 11/03/2025 | 11/28/2025 | 19d | +1.26% |
| Neutral | 12/01/2025 | 12/31/2025 | 22d | +0.84% |
| Defensive LIVE | 01/02/2026 | current | 212d | -4.81% |
Scroll inside the table for the full history. Forward 60-day GC is measured from each run’s start date; a reader cannot trade off a start-date they only identify in hindsight, so this is a transparency log, not an alpha claim. Asterisked (*) returns are partial. The run is less than 60 trading days old. The full 60-day comparable fills in as more data arrives.
What this measures
Platinum (PL) and gold (GC) trade in the broader precious-metals complex. When PGM-complex stress hits, platinum volatility leads gold volatility with a 1-day lag. The engine computes a 1-day-lagged variance spillover coefficient from platinum to gold using rolling window regression of gold returns on lagged platinum squared returns. Elevated spillover indicates platinum volatility is leading gold volatility — a cross-metal stress transmission signal.
Why it matters
Platinum volatility leads gold volatility when PGM-complex stress hits. The 1-day lag captures transmission timing; variance spillover captures its magnitude.
Phase +1 spillover-stress N=621 days delivered mean fwd60d +2.34% versus baseline +1.30%. A 104 basis-point edge across 4,231 trading days.
Variance spillover captures stress but not direction. A spillover-stress regime can coincide with either gold strength (safe-haven channel active) or gold weakness (liquidity unwind dominates).
GO DEEPER
What the Gold Credit-Quality Composite signal actually tells you (and when it does not)
The brief covers the 1-day-lagged variance spillover methodology, three signature episodes (2020-03 COVID panic, 2023-07 calm regime, 2025 moderate windows), and four caveats on reading a cross-metal stress signal.
Signal consumers and siblings
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 05, 2026 11:16 PM ET