Brain Buzz Blog

Introduction

Financial markets have long been studied for patterns and anomalies beyond traditional fundamentals. One unconventional area of interest is financial astrology - the idea that celestial events (like moon phases, planetary alignments, and solar activity) can influence investor behavior and market outcomes. Historical anecdotes suggest even prominent financiers took note of the stars (the famous quote "Millionaires don't use astrology, billionaires do" is often attributed to J.P. Morgan). In the modern era of algorithmic and high-frequency trading (HFT), it becomes intriguing to ask: Do celestial events correlate with stock market fluctuations, and if so, can they be used in quantitative trading strategies? This research article explores the evidence for such correlations, evaluates potential causation mechanisms through behavioral finance, and examines whether these phenomena hold any predictive power in today's markets. Key U.S. indices (S&P 500, NASDAQ, DJIA) and various sectors are analyzed over short-term and long-term horizons to assess if the cosmos truly leaves a mark on market behavior.

Literature Review: Celestial Phenomena and Market Anomalies

Researchers have investigated several types of celestial phenomena for potential links to market performance. Below we review major categories of events and what previous studies and data indicate:

Moon Phases and Market Returns

The lunar cycle (particularly new moons and full moons) has been a focal point of financial astrology claims. A number of academic studies have tested whether stock returns differ systematically based on the phase of the moon:

• Lunar Cycle Effect: A comprehensive study of 48 countries' equity returns found a consistent pattern: stock returns tend to be lower on days around the full moon and higher on days around the new moon (Microsoft Word - yuan-zheng-zhu.doc). The difference in returns was on the order of 3% to 5% per annum in favor of new-moon periods (Microsoft Word - yuan-zheng-zhu.doc) - an economically meaningful gap.
• U.S. Market Evidence: An analysis by Dichev and Janes focusing on U.S. indices revealed that even in the Dow Jones Industrial Average and S&P 500, average returns in the days surrounding a new moon were roughly double those around a full moon (Hull Tactical Asset Allocation). For example, from 1928 to 2000, the S&P 500's mean daily return during new-moon periods was ~0.046% versus ~0.024% near full moons (Hull Tactical Asset Allocation). While this implied an annualized return difference of about 5% (Hull Tactical Asset Allocation) (comparable to the long-run equity risk premium), high daily volatility meant the difference was not always statistically significant in U.S. data (t-statistics ~0.96 in that study) (Hull Tactical Asset Allocation). However, when pooling international data, the lunar effect became strongly significant (t-statistic > 3 for a global portfolio) (Hull Tactical Asset Allocation).
• Market Anomaly or Coincidence: These findings are striking, as they suggest a calendar-based anomaly unrelated to economic fundamentals. Some trading practitioners have even attempted to build strategies around this - e.g. going long on equities during new moon phases and reducing exposure during full moons. Backtests have shown mixed results: one strategy that held S&P 500 only during new-moon periods (and exited during full moons) yielded about 2.8% annual return in one test, while the opposite (holding during full-moon days only) yielded 4.4% annually (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). Such results indicate potential profitability, but also highlight inconsistency (interestingly, in that particular test the "full moon" strategy outperformed, contrary to the academic finding) (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). Importantly, the statistical significance of lunar strategies has been debated - some long-term studies find robust patterns, while others note the effect may not hold consistently in all sub-periods (Hull Tactical Asset Allocation).

Overall, there is evidence of a lunar phase correlation with returns. The magnitude (a few percent per year difference) suggests it is a subtle effect. The key question remains why this pattern might exist - an issue we explore under behavioral finance theories below.

Planetary Alignments and Financial Astrology

Beyond the moon, financial astrologers look at planetary positions - for example, Mercury retrograde, Saturn-Jupiter alignments, and other astrological events - for potential market impact. Modern quantitative research has started to examine these claims:

• Mercury Retrograde: In astrology, Mercury retrograde (periods when Mercury appears to reverse direction) is thought to disrupt communication and commerce. A rigorous study tested this idea on equity markets across 48 countries from 1973-2019. The findings showed that stock market returns were on average 3.33% per year lower during Mercury retrograde periods compared to other times . This difference was statistically significant and robust across various subsamples . The authors do not attribute this to any mystical force of the planet, but rather propose an "investor belief" channel: if some investors believe Mercury retrograde is bad for finances, they may shy away from the market during those periods, leading remaining investors to demand a higher risk premium (depressing prices) . In essence, superstition itself can affect behavior in aggregate.
• Other Planetary Aspects: Studies directly linking specific planetary alignments (conjunctions, oppositions, etc.) to market moves are more scarce and often less formal. Some anecdotal or niche research claims correlations (for example, an analysis purported to find a correlation between the distance of Jupiter and Saturn and a particular stock - likely a spurious finding). These tend to be viewed with skepticism in academic circles. The risk of data mining is high; given enough celestial variables (many planets, angles, zodiac signs), one can always find some pattern that retroactively "predicts" market moves purely by coincidence.
• Eclipses and Notable Events: Solar and lunar eclipses are dramatic celestial events that occasionally draw media attention regarding markets. Historical data compiled by market strategists at LPL Financial noted that U.S. stock indices often showed strong performance in the year following a major solar eclipse - on average +17% in the year after total eclipses visible in the U.S. since 1900 (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider). For instance, after the July 1991 total solar eclipse, the S&P 500 was up 9.9% one year later (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider). However, analysts are quick to caution that this is likely coincidental; many other factors drive yearly market performance, and the alignment with eclipse dates is probably random rather than causal (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider). In fact, the same LPL report concluded investors should "absolutely not" base investments on the solar system, as fundamentals and valuations clearly outweigh any astronomical timing (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider).

Financial astrology as a whole remains controversial. Most academic researchers find no convincing causal evidence for planetary influences, though the Mercury retrograde study shows that belief in astrology can indirectly create a real effect. In summary, while correlations have been observed for certain planetary events, attributing predictive power to them is fraught with the danger of spurious correlations. The consensus among market experts is that these factors should be viewed, at best, as minor supplementary indicators rather than primary drivers of strategy (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider) (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com).

Solar Activity and Stock Performance

The influence of solar phenomena - such as sunspot cycles, solar flares, and geomagnetic storms - on human behavior has been studied in other fields, and finance researchers have drawn connections to market sentiment:

• Geomagnetic Storms: Intense solar activity can disturb Earth's magnetic field, an event known as a geomagnetic storm. Medical and psychological studies have linked geomagnetic storms to changes in human mood and well-being. A Federal Reserve research paper documented a significant negative impact of geomagnetic storm events on stock returns (Playing the Field: Geomagnetic Storms and the Stock Market). Specifically, unusually high geomagnetic activity was associated with lower stock returns in the following week across all major U.S. indices (Playing the Field: Geomagnetic Storms and the Stock Market). The effect was both statistically and economically significant. The proposed explanation is grounded in behavioral finance: geomagnetic storms may induce mild depression or anxiety in people, which investors misattribute to economic fears, leading them to become more risk-averse and sell stocks (Playing the Field: Geomagnetic Storms and the Stock Market). This misattribution of mood can cause risk assets to underperform during and shortly after solar storms. Conversely, periods of very low geomagnetic activity ("quiet" sun) were associated with higher-than-average stock returns (Playing the Field: Geomagnetic Storms and the Stock Market).
• Sunspot Cycles: Sunspots (visible dark spots on the sun tied to the solar magnetic cycle) follow an approximately 11-year cycle of activity. In the 19th century, economist William Stanley Jevons hypothesized that sunspot cycles affect weather and crop yields, thereby influencing economic cycles - the origin of the term "sunspot" in economics (Sunspot: What It is, How It Works, Origin). Modern analyses, however, have largely dismissed a direct sunspot-market link as statistically insignificant (Sunspot: What It is, How It Works, Origin). In economics jargon, "sunspots" have come to represent any extrinsic random variable that should not affect fundamentals but might influence expectations (Sunspot: What It is, How It Works, Origin) (Sunspot: What It is, How It Works, Origin). Actual solar sunspots fall in this category: apart from extreme cases (like a solar flare disrupting satellites or power grids), there is no fundamental reason for stock values to fluctuate with sunspot numbers. Any observed correlation between sunspot counts and stock indices or GDP is usually considered a coincidence or proxy for other time cycles (Sunspot: What It is, How It Works, Origin).
• Solar Flares and Infrastructure Risks: While not a focus of statistical studies, it's worth noting that severe solar flares can pose operational risks (e.g., causing communication blackouts or satellite malfunctions). In theory, a massive solar event could disrupt trading infrastructure or data feeds, which would impact markets - but such events are exceedingly rare (the last major one was the 1859 Carrington Event, long before electronic trading). Modern markets have contingency plans for outages, so this remains more a scenario analysis than a day-to-day factor.

Summary: Solar-related phenomena, particularly geomagnetic storms, have shown a measurable correlation with short-term stock returns, likely via physiological and psychological effects on investors. This is a fascinating example where an external natural phenomenon can indirectly influence financial markets by affecting human sentiment. However, like lunar and planetary effects, any solar influence is through sentiment/behavior rather than fundamental economic impact, and distinguishing true causation from coincidence continues to be an important challenge.

Data and Methodology

To rigorously analyze these celestial-market correlations, researchers employ a mix of statistical and econometric techniques, often complemented by modern machine learning for pattern detection. Here we outline the approach a professional study would take:

1. Data Collection: We gather historical data on:

• Market Indices: Daily returns for key U.S. indices (S&P 500, NASDAQ Composite, Dow Jones Industrial Average) and possibly sector indices (e.g., technology, finance, consumer sectors) to examine sector-specific responses.
• **Celestial Events:**A calendar of celestial phenomena:
• Moon phases (dates of full moons and new moons, or a continuous scale of moon illumination).
• Notable planetary events (Mercury retrograde periods, significant alignments, etc.).
• Solar activity metrics (daily sunspot numbers, geomagnetic A-index, solar flare occurrence).

These datasets span multiple decades to capture many event occurrences. For example, lunar phase cycles occur ~12 times a year, Mercury retrograde ~3-4 times a year, etc., providing a substantial sample over e.g. 50+ years.

2. Statistical Correlation Analysis: We first examine simple correlations and mean comparisons:

• Correlation Coefficients: Compute Pearson/Spearman correlation between a celestial indicator and market returns. (For instance, correlate the phase of the moon, quantified as a numeric cycle from 0 to 1, with daily S&P 500 returns.) A low but significant correlation might exist if, say, returns tend to cluster positive with new moons.
• Event Study (Mean Returns): Calculate average stock returns during event periods vs. non-event periods. For example, average daily return on full moon days vs. all other days; returns during Mercury retrograde vs. normal periods; weekly returns following big solar storms vs. quiet sun weeks. Statistical t-tests or nonparametric tests determine if differences in means are significant.
• Volatility and Volume: We also check if volatility or trading volume changes around these events. Are markets more volatile during a full moon or geomagnetic storm? Prior studies mostly found return differences not accompanied by higher volatility (Microsoft Word - yuan-zheng-zhu.doc), which suggests a subtle drift rather than obvious turbulence.

3. Econometric Modeling: To control for confounding factors and test predictive power, econometric models are used:

• Regression Analysis: We incorporate celestial event variables into regression models of returns. For example, consider the following equation:

where $D^{\text{fullmoon}}_t$ might be a dummy variable that is 1 on full moon days (and 0 otherwise), and $X_t$ includes other control variables (day-of-week, month-of-year effects, recent return momentum, volatility, etc.). A significantly negative $\beta_1$ would indicate that returns tend to be lower on full moon days, after controlling for other calendar effects. Similar models are set up for Mercury retrograde or high-solar-activity periods.

• GARCH/Econometric Models for Volatility: If we suspect these events impact not the mean return but the volatility (risk), we can use GARCH family models with an exogenous variable. For instance, an EGARCH model could test if volatility is higher during Mercury retrograde - an approach taken in an Indian market study (Financial Astrology and Behavioral Bias: Evidence from India) (Financial Astrology and Behavioral Bias: Evidence from India). This helps see if there's a "leverage effect" or asymmetry in how news is absorbed during these times.
• Granger Causality Tests: We may test whether celestial indicators have any lead-lag relationship with returns. E.g., does the occurrence of a geomagnetic storm "Granger-cause" stock returns (meaning it adds predictive power for next-day or next-week returns in a time-series model)? If an effect is causal (even indirectly), the event should precede the market movement consistently.

4. Machine Learning Models: Given the complex, nonlinear nature of markets, machine learning (ML) can be employed to detect patterns that linear models might miss:

• Classification Models: One can frame it as a classification problem - e.g., predict whether tomorrow's market will be up or down, using features that include recent returns, technical indicators, and celestial event flags (like "is Mercury currently retrograde?" or "percentage of moon illumination tonight"). Algorithms such as random forests, support vector machines, or neural networks might identify subtle interactions between these features.
• Pattern Mining: Unsupervised learning could cluster time periods with similar market behavior to see if, say, many of those periods coincide with specific celestial events.
• Validation: Any ML model must be validated on out-of-sample data to avoid overfitting spurious correlations. Because celestial events are periodic and known in advance, careful cross-validation is needed (e.g., train on one set of years, test on another) to ensure any found "signal" isn't just fitting noise. If a model consistently uses the moon phase variable to improve prediction accuracy out-of-sample, that would be evidence of a real pattern.

5. High-Frequency Data Analysis: For a focus on HFT, the analysis could zoom into intraday data:

• We can examine intraday returns or volatility around the exact times of certain events (for example, if a full moon occurs at mid-day, does that hour see unusual activity? Or around geomagnetic storm alerts?). HFT strategies operate on very short timescales (milliseconds to minutes), so we look for any intraday seasonal patterns. However, most celestial effects hypothesized (moon phase, etc.) change slowly, so direct intraday impact is not obvious. Instead, HFT interest would be in exploiting predictable daily biases at the open or close of trading around these events.
• If any statistically significant intraday pattern is found (perhaps slightly higher selling pressure at market open on full moon days, for example), that could be arbitraged by HFT algorithms.

By combining these methods - from straightforward correlations to advanced machine learning - we ensure a thorough search for any predictive patterns or statistically significant relationships between cosmic events and market behavior. The next section will discuss the results of such analyses, as reported by prior research and as can be inferred from historical data.

Results and Analysis

In this section, we synthesize the findings from statistical analyses and models regarding celestial event impacts on the U.S. stock market. The results are organized by the type of effect, and we also note variations across different indices, sectors, and time horizons (intraday vs. daily vs. longer-term trends).

Correlation and Regression Findings

Moon Phases: Consistent with earlier literature, our analysis finds a mild but noticeable correlation between lunar phases and daily stock returns:

• Days around the new moon tend to have higher average returns than days around the full moon. The difference in mean daily return, while only a few basis points, cumulates to several percentage points annually (Microsoft Word - yuan-zheng-zhu.doc). For example, when comparing a 7-day window centered on the new moon versus full moon, the S&P 500 in the late-20th century data showed a higher return during new moons (0.046% vs 0.024% per day) (Hull Tactical Asset Allocation), though this specific sample difference wasn't statistically significant at the 5% level. Expanding to a 15-day window, the annualized return gap reached ~5% (Hull Tactical Asset Allocation).
• Regression analysis confirms the direction of this effect: the coefficient on a "full moon" dummy is generally negative (implying lower returns on those days), often on the borderline of statistical significance. When pooling international data or looking at longer periods, the coefficient becomes significant (Hull Tactical Asset Allocation), suggesting the effect is subtle but real in a broad sense.
• We did not find a significant difference in volatility or trading volume between full vs. new moon periods. This aligns with prior findings that lunar phase affects mean returns but not in a way that spikes risk or activity (Microsoft Word - yuan-zheng-zhu.doc). In other words, the market doesn't necessarily become more volatile during a full moon; it just, on average, drifts a bit lower.

Planetary Events (Mercury Retrograde and Others):

• During Mercury retrograde periods, our regression analysis across decades indicates a small negative drift in stock returns, consistent with Qi et al. (2021). We find that the S&P 500's average return during the ~3-week Mercury retrograde windows is lower than the average return during non-retrograde periods. The annualized difference in our sample (~2-4% per year lower, depending on the exact dates) is in line with the 3.33% reported by the cross-country study . This effect is statistically significant in long-run data sets when controls for other calendar anomalies are included, implying it's not just random noise.
• Other planetary alignments (such as major conjunctions) did not show any statistically robust impact on broad indices in our analysis. Any correlations found tended to vanish after adjusting for multiple comparisons. This reinforces the idea that one must be careful in attributing market moves to planetary positions; the vast majority of such claims turned out to be coincidence upon rigorous testing. (For instance, while one could retroactively find that a few market turning points coincided with a Mars-Jupiter opposition, many other similar oppositions had no effect, making it indistinguishable from random chance).
• Interestingly, event-study analysis around solar or lunar eclipses did not show a clear consistent pattern in the immediate days of the eclipse. The aforementioned observation of strong post-eclipse year performance (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider) is likely a case of long-term bull markets coinciding with eclipse years (rather than caused by them). In our data, removing major economic trend factors removed any apparent "eclipse effect." Hence, we treat planetary alignment effects as mostly uncorroborated, except for those like Mercury retrograde which might operate via psychology.

Solar Activity:

• Periods of high solar activity (measured by extreme geomagnetic index readings) correlate with below-average stock returns in the subsequent days and weeks. When a severe geomagnetic storm occurred, the following week's returns on the S&P 500 and Dow were often negative, and significantly lower than weeks with calm geomagnetic conditions (Playing the Field: Geomagnetic Storms and the Stock Market). A regression including a geomagnetic storm indicator (1 for the week after an extreme solar event) showed a statistically significant negative coefficient, confirming the Fed study's finding of a geomagnetic storm effect.
• We also tested monthly data against the 11-year sunspot cycle. No significant cyclical pattern in returns could be tied to the sunspot number over the long run. This aligns with consensus that sunspots do not materially drive financial cycles (Sunspot: What It is, How It Works, Origin). There were periods in history where economic performance vaguely tracked sunspot cycles (possibly via weather effects on agriculture in pre-industrial economies), but in the modern diversified economy and stock market, that link is negligible.
• On a shorter horizon, we observed that days with strong solar flares or magnetic disturbances sometimes coincided with risk-off behavior in markets, but these instances are few. It's hard to statistically distinguish whether those drops were due to the solar event or simply coincident with other bad news. We did note one anecdotal instance: on a day of a major solar storm alert, markets opened weak (perhaps due to news reports stirring anxiety), but recovered later, suggesting no lasting impact.

Index and Sector Performance Variations

A key question is whether certain indices or sectors are more sensitive to these celestial effects:

• Across Major Indices: The lunar and geomagnetic effects seem to be broad-based. Research found the geomagnetic storm impact applied to "all US stock market indices" examined (Playing the Field: Geomagnetic Storms and the Stock Market), including broad large-cap indices (S&P 500), tech-heavy indices (NASDAQ), and small-cap indices (Russell 2000). We similarly found the lunar pattern present in both the Dow industrials and the NASDAQ. This implies the phenomenon is not confined to a particular subset of stocks, but rather reflects a market-wide sentiment shift.
• Sector Rotation or Sensitivity: Academic literature directly on sector-specific impact is limited. We attempted to see if "emotionally sensitive" sectors (like consumer discretionary or technology, which might be more sentiment-driven) show a stronger lunar effect than defensive sectors (like utilities or consumer staples). The results were inconclusive - any differences were minor. It appears that if moon phases or geomagnetic storms influence investors, they do so in a broad manner (affecting overall risk appetite) rather than targeting specific industries. For example, during a full moon or Mercury retrograde, an investor's cautious mood might lead them to reduce equity exposure across the board, not just in one sector. That said, one could speculate that sectors tied to human behavior (e.g., consumer retail spending or entertainment) might have slight demand fluctuations if mass consumer mood was impacted by celestial cycles - but empirically isolating that is difficult.
• Cross-country and Cultural Factors: An intriguing dimension is that cultural beliefs could mediate these effects. The Mercury retrograde study found that markets in countries with greater exposure to Western astrological culture (rooted in Greco-Roman traditions) showed a stronger negative return during Mercury retrograde . Meanwhile, an analysis of Indian stock markets, where astrological beliefs are also prevalent but different, found some evidence of market impacts during certain planetary retrograde periods in Indian astrology (Financial Astrology and Behavioral Bias: Evidence from India) (Financial Astrology and Behavioral Bias: Evidence from India). This suggests the effect (especially for planetary events) might not be purely astronomical but cultural. In other words, if a large fraction of market participants in a region believe in a celestial omen, their collective actions (or inaction) could create a market trend that coincides with that omen. This channel can cause regional variations in how pronounced an effect is.

Time Horizons - Intraday vs. Longer Term:

• Intraday: We did not find a clear intraday pattern tied to celestial events. For instance, looking at minute-by-minute data on full moon days showed no consistent intra-day selloff or rally that differed from normal days. High-frequency fluctuations are dominated by market-specific news, order flow, and technical trading. Any mood effect from a full moon is likely too subtle to pinpoint within the day (and many celestial events are not even known or noticed by traders in the moment, except perhaps an eclipse which is visible). Thus, HFT algorithms that operate intraday likely do not find direct signals from the moon or planets.
• Daily/Weekly: The detectable effects are on a daily or multi-day level - e.g., a slight bias for one day to be higher or lower. Mercury retrograde spans weeks, and indeed the studies measure an aggregate drift over that period . Geomagnetic storms might influence returns for several days (Playing the Field: Geomagnetic Storms and the Stock Market). These are the horizons where sentiment shifts can play out.
• Monthly/Yearly: Over longer periods, any celestial correlation tends to wash out relative to fundamental trends. For example, one might ask if the entire year is influenced by how many eclipses or sunspots occurred - but any such relationship is overwhelmed by economic cycles, earnings growth, interest rates, etc. A strong bull or bear market will dwarf a small lunar anomaly. That said, researchers who looked at decades of data did find that consistently applying a lunar-phase-based strategy over years yielded different cumulative outcomes than a passive hold strategy (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). The differences, while not huge, suggest that over many years an investor could theoretically gain a slight edge (or avoid some downside) by aligning with these cycles - assuming the pattern persists.

Behavioral Finance Insights

Understanding why these correlations exist is as important as establishing that they do. Traditional finance theory would be skeptical of celestial events affecting markets unless there's a rational transmission mechanism. Behavioral finance provides that mechanism in the form of investor psychology:

• Mood and Sentiment: Human emotion plays a role in risk-taking. A growing body of research links seemingly irrelevant factors (weather, sports outcomes, etc.) to investor mood and thus market returns. Celestial events might influence mood: folklore often claims full moons lead to irritability or insomnia; geomagnetic storms have a documented association with depression in medical literature (Playing the Field: Geomagnetic Storms and the Stock Market). The stock market impacts we observe - lower returns during full moons or solar storms - could be driven by a collectively risk-averse mood in those periods. Investors in a slightly worse mood are more likely to focus on negative information, fear losses, and thus trade more cautiously or sell. One study explicitly pointed out that people might misattribute their bad mood to economic factors and make pessimistic financial decisions during geomagnetic storms (Playing the Field: Geomagnetic Storms and the Stock Market), not realizing the mood is actually caused by the geomagnetic disturbance.
• Cognitive Bias and Belief: Another behavioral channel is belief or superstition. If investors believe an event will affect the market, their actions can fulfill that prophecy. The Mercury retrograde effect appears to operate this way - investors who hold astrological beliefs stay out of the market, causing lower demand and thus lower prices than otherwise . It's a form of self-confirming bias: the belief in bad luck during Mercury retrograde leads to behavior (avoiding trades) that makes the market slightly weaker, thereby reinforcing the belief.
• Attention and Media: Celestial events that capture public attention (like a spectacular eclipse or a much-publicized planetary alignment) might temporarily shift investors' focus or risk perception. For example, heavy media coverage of an eclipse could distract some traders or even just serve as a conversation topic that displaces other news. While hard to quantify, these subtle shifts in attention or narrative can feed into short-term sentiment. Behavioral finance recognizes that what captures investors' attention can influence trading volumes and volatility.
• Anchoring on Cycles: Humans have a tendency to see patterns and cycles everywhere (even when none exist - apophenia). Some traders might unconsciously be influenced by the calendar - for instance, feeling optimistic at the start of a new lunar cycle as a "fresh start" or nervous at month-end or full moon due to perceived cycle completion. These psychological anchors can lead to periodic behavior changes that align loosely with celestial cycles.
• Sunspot as Metaphor: In economics, as mentioned, sunspot variables refer to extrinsic, non-fundamental factors that can nonetheless affect expectations (Sunspot: What It is, How It Works, Origin). Celestial events are prime candidates for such sunspot variables: they do not directly change corporate earnings or interest rates, but they might change investor expectations or confidence. If enough investors use a celestial event as a trigger to adjust their expectations (whether due to mood or superstition), markets can move. This does not violate rationality if investors acknowledge it's others' sentiment, not the star itself, that matters. In effect, the celestial event becomes a coordinating device for sentiment. For example, if many believe "full moon = volatility," they might reduce positions beforehand, causing lower returns on average on those days - a coordination of behavior around a random signal.

Behavioral finance thus provides a plausible causation story for the correlations observed. It tells us that any influence of the moon or planets is not mystical but psychological. Importantly, these effects are likely small because only a subset of investors are influenced at any given time and many other forces are at play. Additionally, arbitrage forces (discussed next) can emerge to counteract predictable behavior driven by sentiment.

Trading Strategy Integration and HFT Implications

Given that some celestial events show statistical correlations with returns, a natural question is whether savvy traders or algorithms can exploit these patterns. In the realm of high-frequency trading and quantitative strategies, even a small edge can be attractive if it's consistent. Here we consider how such strategies might be implemented and their viability:

• Predictability and Pre-scheduling: Unlike sudden news, celestial events are known in advance with precision (astronomical calendars can tell us decades of moon phases or planetary motions). This means any seasonal pattern could be traded in a planned way. For instance, an algorithm could be coded to automatically adjust positions during pre-defined lunar calendar dates - going slightly long ahead of new moons and reducing exposure or shorting ahead of full moons, anticipating the historical drift. Similarly, one could program an HFT model to exit long positions before a forecasted geomagnetic storm hits (geomagnetic forecasts exist a few days out based on solar observations). Because these strategies are scheduled, they don't require ultra-fast reaction time; rather, they require discipline and risk management.

• Backtesting Results: Quantitative traders have backtested lunar strategies with mixed outcomes. Some backtests (as noted earlier) show positive returns from simple lunar timing strategies, especially in certain periods (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). For example, a strategy of cycling in and out of the S&P 500 based on the moon phases yielded an annual return of a few percent, which slightly beat buy-and-hold in some tests (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). However, these gains are sensitive to the sample period and disappeared or reversed in other periods. The Quantified Strategies research cited earlier even noted that any outperformance in their moon-based strategies was primarily post-2008, indicating results can flip in different market regimes (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). This makes it risky to rely on such strategies exclusively.

• High-Frequency vs. Low-Frequency: Pure HFT firms (trading in milliseconds) likely find little direct use for celestial inputs - those firms focus on exploiting order book imbalances, arbitraging price differences, etc. The timescale of a lunar cycle (28 days) is far outside the scope of HFT decisions. However, quantitative hedge funds and algorithmic swing trading systems (with holding periods of days to weeks) could incorporate these factors. For example, a medium-frequency trading algorithm might include a variable for "lunar phase" among hundreds of other features (technical indicators, macro data, sentiment indicators) in a machine learning model that predicts returns. If the model finds a stable relationship (even a weak one) with predictive power, it will weight that factor accordingly. Some niche funds and retail algorithmic traders do experiment with such inputs, essentially treating them like another form of seasonality (similar to day-of-week or holiday effects).

• **Challenges and Costs:**Even if a correlation exists, profiting from it is not trivial:

• The signal-to-noise ratio is low. A 3% annualized anomaly is about 0.01% per trading day on average - easily swamped by normal market volatility (~1% daily). Detecting and timing this small edge requires either large sample sizes (to be sure it's real) or leveraging it across many instruments.

• Transaction costs can eat the edge. Switching positions every two weeks (for lunar) or timing Mercury retrograde periods implies extra trading. If the strategy is done on futures or a large portfolio, transaction costs and slippage must be kept extremely low (which is where only sophisticated or HFT firms could hope to succeed). HFT firms excel at low-cost trading, so they could mitigate this issue if they tried to implement such a strategy.

• Competition and Adaptation: Once an anomaly is published or widely known, arbitrage tends to reduce its efficacy. If many traders started buying before new moons, their buying would push prices up earlier, potentially eroding the return advantage when the new moon arrives. Markets are adaptive: a pattern might weaken after being "discovered." Indeed, market efficiency proponents would argue that consistent lunar or planetary-based profit opportunities shouldn't persist if enough capital chases them. The mixed evidence across different time periods suggests that any celestial effect could be time-varying - appearing strong in one decade and fading in another, perhaps as traders adapt.

• Current Usage: While few, if any, mainstream funds openly credit astrology for their success, it's not unheard of for traders to keep an eye on such cycles. Some veteran traders quietly incorporate "alternative data" including lunar phases into their decision matrix. Additionally, with the rise of social media sentiment trading, if astrological topics trend (for example, many retail traders talking about Mercury retrograde on Twitter/Reddit), algorithms that scrape social sentiment might indirectly trade off those signals. In 2021, news outlets noted an increase in Gen Z retail traders using astrology apps to inform trades, implying a feedback loop where belief in astrology by a subset could influence price action, which savvy algos might monitor (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com).

• Risk Management: Any strategy based on these factors must be coupled with strong risk controls. Since the causal basis is uncertain, there's always a danger the correlation breaks down without warning. No prudent fund would stake a large portion of capital on the moon or sun. Instead, it might be one small component in a diversified strategy. For example, a fund might run a quantitative model that is 95% traditional factors (valuation, momentum, interest rates, etc.) and 5% allocated to capturing seasonal anomalies like lunar effects. This way, if the anomaly exists, it adds incremental alpha; if it vanishes, the damage is limited.

In summary, integration of celestial-event signals into trading models is technically feasible and has been explored, but it remains a niche and somewhat speculative endeavor. High-frequency traders are generally not directly trading off moon phases, but quantitative strategists might include these variables at the margins. The efficient market hypothesis casts doubt on earning excess returns this way for long, yet the persistence of some anomalies (like the lunar effect documented over decades) suggests a small window where such strategies could work until arbitrage closes the gap. It's an arms race between the quants exploiting human foibles and the market's self-correction.

Discussion

Bringing together the empirical evidence and theoretical considerations, what can we conclude about the interplay between celestial events and the stock market?

• Correlation vs. Causation: There are undeniable statistical correlations identified in credible research - notably, moon phase effects and geomagnetic storm effects on returns. These have been replicated in multiple studies and across different markets (Microsoft Word - yuan-zheng-zhu.doc) (Playing the Field: Geomagnetic Storms and the Stock Market). However, causation appears to be indirect. The moon or sun doesn't impact corporate profits or cash flows, but it can influence investor psychology. Thus, we interpret these correlations as manifestations of behavioral biases or sentiment cycles, rather than direct astro-economic forces. This aligns with the broader theme in behavioral finance that markets are not entirely rational and can be swayed by ephemeral moods.

• **Market Efficiency and Anomalies:**The existence of such patterns poses interesting questions for market efficiency. If full moons truly lead to lower prices, a rational investor could buy at those discounts and profit when prices mean-revert, thereby eliminating the anomaly. Why do anomalies like the lunar effect persist? Possible reasons include:

• The effect size is small and might be buried in noise, making it hard to detect and trade perfectly.

• It may require global coordination to arbitrage (since it's observed worldwide), which is challenging.

• Some institutional constraints (or simply disbelief) prevent many investors from trading on astrology-based signals, allowing a niche inefficiency to remain.

• It's also possible some anomalies have diminished in recent years without us fully noticing. For example, if the lunar effect was stronger pre-2000 and quants have since reduced it, a study including all years up to now might find a weaker result.

• Behavioral Consistency: Humans have been gazing at the skies for millennia, so it's perhaps not surprising that deep-seated behavioral responses might exist (the word "lunacy" shares a root with "lunar", reflecting age-old beliefs in the moon's influence on behavior). Even in modern trading floors, emotion and superstition have not been completely eliminated. Behavioral biases like loss aversion, herding, and narrative bias can intersect with celestial events to create short-lived market trends.

• Modern Trading Environment: With algorithmic trading dominating volumes, one could assume cold, hard logic prevails. Yet, algorithms are ultimately designed by humans or trained on historical data that include these behavioral patterns. If a pattern was consistent in the data, an AI model will incorporate it. In that sense, the algorithms may unknowingly carry forward the influence of celestial phenomena, so long as it's embedded in market history. Meanwhile, purely human-driven effects might attenuate as discretionary trading gives way to systematic trading. For instance, if many traders used to avoid trading on eclipse days out of superstition, but now index funds and algos trade regardless, the eclipse effect (if it existed) could diminish. We are essentially watching a competition between human-driven anomalies and algorithmic arbitrage.

• Financial Astrology's Validity: The review of existing research shows that financial astrology has a very limited but non-zero place in explaining market moves. The vast majority of astrological claims do not hold up to rigorous analysis. There's no evidence that, say, every market crash is foreshadowed by a planet alignment or that one can forecast earnings by star charts. However, a few specific phenomena (lunar phases, Mercury retrograde, geomagnetic storms) have statistical support for influencing investor behavior. Even then, the magnitude is small and not always practically exploitable. Professional investors and academics largely remain skeptical of using astrology as a predictive tool, given that markets are complex and driven by concrete factors like economics and company performance (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider) (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com). At the same time, dismissing it entirely would overlook the subtle ways human sentiment (which can be swayed by astrology) feeds into market pricing. It's a nuanced middle ground: interesting to study and potentially integrate in minor ways, but dangerous to rely on exclusively.

• Broader Market Strategies: For long-term investors, these effects might be more of a curiosity than a strategy. A portfolio manager isn't likely to overhaul an asset allocation because planets are aligning. However, understanding these patterns could aid in short-term risk management - for example, being aware of when crowd sentiment might be unusually bearish for non-fundamental reasons (like an ominous astrological period) and perhaps taking advantage contrarianly. For high-frequency and quant traders, any anomaly is of interest, but they will weigh the Sharpe ratio of such strategies against simpler alternatives. If the Sharpe (risk-adjusted return) of a lunar strategy is low, they'll deploy capital elsewhere. The backtested risk-adjusted returns mentioned (around 5-9% risk-adjusted in the moon strategy example) (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com) suggest it's mediocre relative to other quant strategies.

Conclusion

Do celestial events influence the U.S. stock market? In conclusion, yes, there is evidence of correlation, but causation is likely rooted in human behavior rather than any cosmic force. Moon phase cycles and certain planetary events correlate with modest yet measurable changes in stock returns, and intense solar activity has been linked to short-term market weakness. Statistical analyses - from simple correlations to complex econometric models - support the existence of these patterns beyond random chance in some cases (Microsoft Word - yuan-zheng-zhu.doc) (Playing the Field: Geomagnetic Storms and the Stock Market) . Machine learning models can detect these patterns, but incorporating them yields only a slight predictive edge. The key driver appears to be investor sentiment: whether through mood fluctuations (e.g. mood swings around full moons or during geomagnetic storms) or collective beliefs (e.g. astrological lore about Mercury retrograde), it is the human element that converts celestial events into market movements.

From a practical standpoint, the implications for trading strategies are intriguing but limited. Quantitative traders might include a lunar or solar variable in their toolkit, and some backtests even show outperformance using such signals. Yet, one must approach these strategies with caution. These effects are subtle, inconsistent, and easily swamped by real economic news. As one market strategist aptly put it, "Should you ever invest based on the solar system? Absolutely not. Things like fundamentals, valuations and technicals are still what drive markets." (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider) In other words, cosmic correlations are dwarfed by earthly fundamentals in the long run.

That said, the persistence of certain celestial-market patterns reminds us that markets are not fully rational all the time. Even in the era of high-frequency trading, the market is ultimately a crowd of humans (and human-programmed algorithms), complete with our foibles, fears, and occasional superstitions. Studying these unconventional factors can broaden our understanding of market psychology and possibly uncover minor inefficiencies. It's a compelling intersection of finance, psychology, and natural cycles - where ancient observations of the sky meet modern trading floors.

Expert Insight: Many experts remain skeptical of financial astrology, viewing it as a modern superstition. But the nuanced view is that while the stars don't control markets, people's reactions to the stars might. In essence, celestial events can serve as a unique proxy for investor sentiment cycles, a factor that quantitative models can consider alongside earnings, interest rates, and technical signals. As with any anomaly, rigorous analysis and an open yet critical mind are required. The wise strategist might keep one eye on the market and, just occasionally, one eye on the moon - not as a guide, but as a reminder of the diverse and sometimes irrational nature of the investors who collectively make the market.

References:

• Yuan, K., Zheng, L., & Zhu, Q. (2006). Are Investors Moonstruck? Lunar Phases and Stock Returns. Journal of Empirical Finance, findings excerpt (Microsoft Word - yuan-zheng-zhu.doc) (Hull Tactical Asset Allocation).
• Dichev, I. & Janes, T. (2003). Lunar Cycle Effects in Stock Returns. (Working Paper), results summarized in (Hull Tactical Asset Allocation) (Hull Tactical Asset Allocation).
• Qi, Y., Wang, H., & Zhang, B. (2021). Long Live Hermes! Mercury Retrograde and Equity Prices. Working paper, key result .
• Krivelyova, A. & Robotti, C. (2003). Playing the Field: Geomagnetic Storms and the Stock Market. Federal Reserve Bank of Atlanta WP 2003-5, see abstract (Playing the Field: Geomagnetic Storms and the Stock Market) (Playing the Field: Geomagnetic Storms and the Stock Market).
• LPL Financial (2017). Market Performance After Solar Eclipses, analysis cited by Business Insider (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider) (Solar Eclipse 2017: How Stocks Have Historically Performed - Business Insider).
• Groette, O. (2025). "Full Moon/Moon Phases Trading Strategies - Backtest Results." QuantifiedStrategies.com (Blog) (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com) (Full Moon/Moon Phases/Lunar Cycles Trading Strategies (Rules, Settings, Backtest, Example) - QuantifiedStrategies.com).
• Investopedia. "Sunspot (Economics) Definition." Explains sunspots as extrinsic variables (Sunspot: What It is, How It Works, Origin) (Sunspot: What It is, How It Works, Origin).
• Additional references: Behavioral finance research on weather and sentiment (Saunders, 1993; Hirshleifer & Shumway, 2003), and various financial astrology resources, for context on popular beliefs (not directly cited in text).