Project 02: Planetary Cycles and Market Volatility #

Source: bigastrologybook.com/2/research/19/project-2 Archive Date: 2026-03-21 Book: The Big Astrology Book of Research by Renay Oshop Dataset: 18,869 trading days (1950–2024) — Yahoo Finance + Swiss Ephemeris

Research Question #

Is there a measurable, statistically robust correlation between major planetary cycles — particularly Jupiter-Saturn — and changes in stock market volatility or returns?

Hypothesis #

Major planetary cycles, especially the ~20-year Jupiter-Saturn synodic cycle, show statistically significant correlation with economic indicators beyond what would be expected by chance.

Why This Question Matters #

Financial astrology is one of the oldest practical applications of astrological thinking, and one of the most testable. Unlike personality traits or life outcomes, market data is publicly available, precisely timestamped, and covers decades. If planetary cycles encode any real structural information about human collective behavior, markets — driven entirely by human psychology — are where a signal should be detectable.

This project applied the full toolkit of modern time-series econometrics to find out.

Data #

Data is real, not synthetic. No concerns about provenance.

Methods #

• ARIMA — time-series modeling and out-of-sample forecasting
• GARCH(1,1) — volatility modeling that accounts for clustering (calm periods followed by turbulent ones)
• Granger Causality — tests whether past planetary positions predict future market prices
• Bootstrapping (N=1,000) — resampling to construct confidence intervals that don't depend on normality assumptions
• Polar Cycle Analysis — market returns and volatility mapped to the full 360° of the Jupiter-Saturn cycle
• 15-Pair Expanded Analysis — Mars, Jupiter, Saturn, Uranus, Neptune, Pluto in all combinations

Results #

1. The Core Finding: Real But Tiny #

The headline statistic is a Pearson correlation of r = 0.0226 (p = 0.0019) between Jupiter-Saturn orb degrees and GARCH-derived market volatility.

That p-value is real. Bootstrapping on 1,000 resamplings produced a 95% CI of [0.0050, 0.0354] — the interval does not cross zero, meaning this is almost certainly not noise. The correlation is statistically genuine.

But the practical meaning is essentially nil: planetary data explains less than 0.1% of market variance. For context, the day of the week explains more.

Directional surprise: The correlation is positive with orb degrees — meaning as Jupiter and Saturn move further apart, volatility increases. Exact aspects (0° orb) are associated with calmer markets. This is the opposite of what most financial astrology would predict.

2. Predictive Power: Worse Than Nothing #

Bottom line: Adding planetary data to a predictive model makes it slightly worse. Past planetary positions do not help forecast future prices. The signal, while statistically detectable, carries no actionable predictive information.

3. The Trine Instability Paradox #

The polar cycle analysis — mapping all 74 years of data to the 360° wheel of the Jupiter-Saturn cycle — produced the project's most surprising finding.

The Trine (120°) is the most dangerous zone in the data:

The Trine is traditionally considered astrology's most "harmonious" aspect — the one associated with ease, luck, and flow. In market data, it correlates with sharp drops and high instability. The Conjunction, meanwhile, generates comparable turbulence but in a growth direction — a "market reset" rather than a correction.

The highest actual returns appear at minor aspects (Quincunx and Semi-Square) that most astrological systems treat as insignificant.

One possible astrological interpretation: the "ease of flow" of a Trine may manifest in markets as "ease of selling" — capitulation and drawdowns happening without friction. This is speculative, but worth exploring.

4. Expanded Analysis: 15 Planetary Pairs #

The single Jupiter-Saturn pair analysis was expanded to all combinations of Mars, Jupiter, Saturn, Uranus, Neptune, and Pluto. Key findings:

The Conjunction Effect: The 0° zone (355°–5°) appears as a primary volatility peak in 7 of 15 pairs — roughly 50%. This "New Moon" pattern — cycle starts triggering market instability — is the most consistent structural finding across the entire dataset.

Multiple testing caveat: With 15 planetary pairs, multiple aspects, and multiple metrics, the number of hypothesis tests is large. A Bonferroni or FDR correction would tighten the significance thresholds. The Saturn-Uranus result (r = −0.18) is robust enough to survive correction; the weaker trace signals (+0.03 to +0.05) should be treated as suggestive rather than confirmed.

The Four Structural Findings #

I. The Conjunction "Heartbeat" #

Volatility spikes at 0° (synodic cycle start) for half of all tested planetary pairs. This is the most replicable finding. It aligns with the "New Moon" archetype: the start of a cycle is a period of instability, but not necessarily decline.

II. The Saturn-Uranus "Disruption" Signal #

With r = −0.18, this is the strongest market-astrology correlation in the dataset — nearly 8× stronger than Jupiter-Saturn. Volatility spikes at hard aspects (0°, 90°) as expected, but also at the Trine (120°), suggesting this pair's influence is more pervasive than standard astrological frameworks predict.

III. The Trine Anomaly #

In the Jupiter-Saturn cycle, the 120° aspect consistently correlates with negative returns and high volatility. This directly contradicts the standard astrological interpretation of the Trine as benefic. Whatever mechanism might underlie planetary-market correlation, it does not map onto traditional astrological symbolism in a straightforward way.

IV. Bullish vs. Bearish Volatility #

Not all volatility is equal. The polar analysis distinguishes:

• Bullish volatility — high activity at Conjunctions (~0°), correlated with positive returns
• Bearish volatility — high activity at Trines (~120°), correlated with negative returns

This is a conceptually useful distinction if the framework is treated as a risk environment model rather than a price predictor.

Conclusion #

Seventy-four years of daily market data and rigorous econometric testing produce an honest answer: planetary cycles are statistically detectable in market behavior but practically useless for prediction.

The correlation is real — bootstrapping confirms it. It's also vanishingly small. ARIMA and Granger tests make clear that no predictive model benefits from including planetary data; it adds noise. The p-value (0.0019) reflects the size of the dataset (18,869 days) more than the size of the effect (<0.1% of variance explained).

What the data does support is a more modest claim: planetary cycles may function as background "seasons" of volatility — describing the risk climate rather than forecasting daily prices. The Conjunction Effect and Saturn-Uranus signal are the best candidates for further investigation.

The genuinely surprising result — that the Trine, astrology's "lucky" aspect, is associated with market drops and instability — is the most intellectually interesting finding here. It suggests that if any planetary influence operates, it does so by its own logic, not according to the symbolic tradition.

This project does not constitute financial advice. Statistical correlations in historical data do not guarantee future performance.

Archived code and raw data outputs preserved in backup/.

Planetary Cycle Correlation with Economic Indicators #

Research Question #

Is there a measurable correlation between major planetary cycles (particularly Jupiter-Saturn conjunctions) and market trends or economic indicators?

Hypothesis #

Major planetary cycles, especially Jupiter-Saturn conjunctions, show statistically significant correlation with changes in economic indicators beyond what would be expected by chance.

Background #

Astrologers have long claimed that Jupiter-Saturn cycles (~20 years) correspond to major economic shifts. This research applies rigorous time-series analysis to test whether such correlations exist in historical data, while carefully controlling for confounding factors.

Data Sources #

• Historical stock market data: S&P 500, Dow Jones from Yahoo Finance or FRED
• Planetary ephemerides: Swiss Ephemeris for precise planetary positions
• Economic indicators: GDP, unemployment, inflation from World Bank or BLS
• Historical commodity prices: Gold, oil from public databases

Mathematical Methods #

1. Cross-correlation analysis: To detect lagged relationships between variables
2. Fourier transforms: To decompose cyclical patterns in both datasets
3. ARIMA models: For time-series forecasting and residual analysis
4. Granger causality tests: To assess predictive relationships
5. GARCH models: For volatility modeling
6. Polar Cycle Analysis: Mapping market returns and volatility to 360-degree planetary cycles.
7. Bootstrapping: Resampling 1000x to establish confidence intervals.

Implementation Plan #

Step 1: Data Collection #

• Download S&P 500 daily data (1950-present)
• Generate Jupiter-Saturn positions and aspects over same period
• Collect additional economic indicators (GDP, unemployment)

Step 2: Data Preprocessing #

• Align all time series to common frequency (daily/monthly)
• Calculate planetary aspects and orbs
• Create indicator variables for major configurations
• Log-transform financial data as needed

Step 3: Exploratory Analysis #

• Plot time series with planetary cycle overlays
• Calculate basic correlations
• Identify potential lead/lag relationships

Step 4: Statistical Modeling #

• Apply cross-correlation analysis
• Perform Fourier analysis on both series
• Fit ARIMA models and test residuals
• Run Granger causality tests
• Build GARCH models for volatility analysis

Step 5: Validation #

• Use out-of-sample testing
• Bootstrap confidence intervals
• Compare effect sizes to known market factors

Expected Outputs #

• Correlation matrices between planetary cycles and economic indicators
• Spectral analysis showing common frequencies
• Statistical tests for predictive relationships
• Clear visualization of any detected patterns

Required Python Libraries #

pyswisseph
pandas
numpy
scipy
statsmodels
arch (for GARCH)
matplotlib
yfinance
fredapi

Ethical Considerations #

• Clearly state this is not financial advice
• Report all tested hypotheses (avoid p-hacking)
• Acknowledge limitations and alternative explanations

Data Provenance #

Economic Indicators #

• Source: Federal Reserve Economic Data (FRED).
• Link: https://fred.stlouisfed.org
• Publisher: Federal Reserve Bank of St. Louis.