Coherence Analytics explores whether random systems show structured deviations around moments of heightened human attention, shared emotional significance, and collective events. It does not assume the answer. Instead, the platform gives researchers and the public tools to explore, compare, and test these patterns across global events, individual devices, and model-based applications.
You do not need to begin with the whole theory. Start with the level that fits your question: a global event, a single device, a model comparison, or a structured hypothesis.
Use Explorer to examine coherence peaks and compare them with candidate events. Read it as a context-building tool, not as proof that any event caused a signal.
Use Device Data for local inspection: one device, one period, one comparison sample, and one statistical question at a time.
Use market model forecasts to compare. The question is whether coherence-derived variables sometimes add information beyond conventional inputs.
Use Hypothesis when a pattern is interesting enough to formalize. The goal is to make assumptions explicit and testing reproducible.
Coherence Analytics is experimental and research-oriented. It supports exploration, model comparison, and hypothesis generation. It does not provide investment advice, trading recommendations, or proof of causality.
This section compares a market-only baseline with a GCP2-augmented model. Model A uses conventional market inputs. Model B adds the prior-day GCP2 signal input, using the same naming logic as Explorer. The Estimated State shows the current HMM market regime. Click for method notes.
↓The page compares two otherwise similar model structures. Model A uses conventional market information only. Model B adds a GCP2-derived signal input to test whether it contains incremental information under some market conditions. The signal names match Explorer: Peak deviation for sharp 15-minute spikes, and Cumulative drift for day-long build-up.
Earlier peer-reviewed work found statistical relationships between GCP-derived signal measures, equity returns, and VIX-related market stress. This page treats that relationship as an empirical modelling question: does adding a GCP2 signal input improve, weaken, or leave unchanged the model output relative to the market-only baseline?
The HMM regime layer classifies market conditions as Calm, Noisy Stress, or Coherent Stress. The purpose is to test whether the coherence-augmented model behaves differently across market states rather than assuming one stable effect at all times.
The ORB Proxy is presented as an alternative intraday information metric. Whereas Model B reports a GCP2-conditioned forecast using the prior-day model signal input, the ORB Proxy tracks how the market itself begins to resolve information during the session.
This measure draws on earlier work by Ulf Holmberg with Carl Lönnbark and Christian Lundström (2013, Finance Research Letters), where opening-range dynamics were treated as statistically informative rather than merely descriptive.
Read together, the two indicators offer complementary views of the same trading day: Model B captures a forecast shaped by prior GCP2 signal conditions, while the ORB Proxy provides an intraday market-confirmation metric based on observed price discovery.
| Date | Return | Close |
|---|---|---|
| Loading… | ||
The model outputs shown here are experimental research results for comparison and hypothesis generation. They are not investment advice, trading recommendations, or proof of causality.
–
This explorer lets you inspect days when coherence signals rise and compare them with independently documented events, maps, timing, and regional context. Read more ↓
The Explorer starts from a simple question: when the GCP2 network shows an unusual coherence pattern, what was happening in the world around that time?
The tool combines GCP2 network data with external sources such as Wikipedia, GDELT, USGS, and calendar-based event data. The goal is to support structured exploration and better hypotheses, not to claim that any single event caused a signal.
Select any day in the calendar. The platform identifies peak coherence intervals in the GCP2 network and surfaces events that align in time and context, based on predefined criteria such as timing precision, regional relevance, and independent data sources. The result is a structured record designed for exploration, comparison, and hypothesis generation.
Periods of elevated stress, volatility, or disruption provide natural test cases for coherence signals. The platform evaluates whether such conditions align with structured deviations under controlled comparisons.
Some events draw unusually broad and synchronized attention. The platform explores whether coherence-derived measures change during such periods, and how those changes compare with external event data.
A distributed network makes it possible to examine whether deviations appear locally, regionally, or globally. The focus is on patterns across locations, time scales, and event contexts.
Collective experiences can involve grief, relief, celebration, fear, or unity. The platform examines whether different classes of shared events are associated with distinct statistical patterns.
This explorer highlights moments when global coherence signals rise — and shows what was happening in the world at the same time. Click to read more.
A global network of random number generators continuously produces random data. Occasionally, the combined signal shows small deviations from expected behavior. These deviations are measured as coherence.
The question this platform explores is simple: when coherence rises, what was happening in the world at that time?
For each day, the platform identifies the moment when coherence is highest, then looks for events that occurred at the same time using independent data sources such as Wikipedia activity, global event databases, and seismic records.
The explorer does not assume that events cause the signal, or that the signal predicts events. It provides a structured way to examine whether consistent patterns appear when coherence is evaluated alongside real-world timelines.
Device Data lets you inspect a single public GCP2 device in a more concrete way than the network-level tools. Choose a device, define a repeated daily window or date range, and test whether that subset differs from a chosen baseline.
If you are new here, start with a repeated daily GCP2 timestamp span and compare it with the rest of the sample. A 60-minute span is a simple first test.
This tool checks whether one device's coherence values look systematically different in a repeated intraday window, a full calendar month, or a custom date period than they do in a chosen baseline. That baseline can be the full sample, the full sample with the subset removed, or, in custom-period mode, a second standardized comparison period using the same intraday window on each day.
How it works. The platform starts from the timestamped public GCP2 device-coherence export, converts timestamps to your selected time zone, and selects rows whose coherence timestamp falls inside the chosen window. Each device_coherence row is interpreted as a 60-minute trailing summary ending at that timestamp. The tool then computes one daily mean from the selected rows and compares those day-level means against the chosen baseline using Welch's t-test.
If you choose 09:30 for 30 minutes, the test selects GCP2 rows timestamped after 09:30 and up to 10:00 local time. Because each selected value summarizes approximately the previous 60 minutes, the result should be interpreted as a lagged coherence signature, not as raw data only from 09:30-10:00.
Why this can be worth testing. Research in the RNG and Global Consciousness literature suggests that if departures from expectation exist, they are usually small and may only become statistically visible when repeated observations are aggregated over time. The idea is therefore not that one unusual timestamp proves anything, but that a stable difference may become detectable when the same window is examined across many days.
How to interpret a result. A low p-value means the selected window differs statistically from the chosen baseline under this model. It does not by itself identify the cause. A difference could reflect operational routines, local environmental factors, date-specific conditions, attention-related clustering, or ordinary noise. The result should be treated as a flag for follow-up, not as proof of mechanism.
-
This section is not meant to prove the platform’s assumptions in one step. It shows the research path behind the tools: first the theoretical question, then the empirical evidence from random systems, and finally the applied question of whether coherence-related measures contain useful information in market and event contexts.
Conceptual models that motivate why random number generators have been used in consciousness-related research: informational-constraint models, entropy-based accounts, decision-selection models, quantum/operator-machine theories, and broader views in which consciousness may not be treated as purely epiphenomenal.
The core theoretical issue is whether cognitive or conscious states can be associated with structured deviations in probabilistic systems. CESM is the platform's primary working model, but it is presented here alongside broader discussions of quantum observation, entropy, anomalous cognition, decision selection, and non-reductive accounts of consciousness.
Introduces the Cognitive Entropy Shift Model (CESM), the platform's main theoretical framework. CESM proposes that consciousness can function as an informational constraint on probabilistic systems, with passive emotional attention and goal-directed intention producing different effects. The model predicts entropy-linked deviations in physical RNG output and is tested directly in a two-year controlled RNG study.
DOI: 10.31275/20263735 →Presents a physics-journal framework in which consciousness is treated not as an emergent product of neural activity, but as a foundational field from which space, time, matter, and individual awareness arise. The model connects universal consciousness with quantum field theory, symmetry breaking, quantum fluctuations, and emergent space-time, providing a high-level ontological bridge for non-local consciousness models.
DOI: 10.1063/5.0290984 →Analyzes the full GCP dataset through an entropy lens, asking whether long-run deviations in random data contain information-theoretic structure beyond classical event-based analyses. This is directly relevant to CESM because it frames GCP anomalies as entropy-linked deviations rather than merely isolated statistical excursions.
DOI: 10.3390/e24121698 →Presents Decision Augmentation Theory, an alternative informational explanation in which anomalous effects arise because human decisions are subtly biased by anomalous cognition, rather than because consciousness directly affects random physical systems. CESM explicitly belongs in this discussion because it differs from DAT by predicting entropy-linked deviations tied to distinct cognitive states.
Archived full text →The core PEAR theoretical synthesis, treating consciousness as a measurable variable in operator-machine systems. It provides the engineering-style framework in which small, repeated deviations from chance are treated as physical anomalies requiring statistical characterization rather than anecdotal interpretation.
ICRL Press →Schmidt's theoretical synthesis of nearly two decades of REG research, proposing a quantum-mechanical model in which conscious observation biases probability distributions. It is the clearest early theoretical ancestor of the later operator-machine, GCP, and CESM frameworks.
Full text (PDF) →The empirical record on whether physical random systems deviate from chance under human attention, intention, or collective events — from recent GCP synthesis backward to the original laboratory RNG work.
These studies provide the empirical basis for asking whether consciousness-related variables can be detected in random physical systems: first in controlled laboratory REG studies, then in distributed global RNG networks during major collective events.
A recent synthesis of the full GCP programme, summarizing 17 years and 500 pre-registered event replications with a compounded result of Z = 7.31. Also discusses the market simulation studies connected to this platform, making it an important external recognition of the practical relevance of GCP data.
DOI: 10.31156/jaex.25553 →Provides external validation of the GCP-market link by showing that GCP1 Max[Z] co-varies with Google search volumes for emotionally absorbing events. This supports the interpretation of Max[Z] as a collective-attention proxy rather than a generic news-flow measure.
DOI: 10.1016/j.explore.2022.07.007 →The landmark GCP paper. A global network of physical RNGs showed statistically significant deviations during major world events such as 9/11, the death of Princess Diana, and New Year celebrations. This paper establishes the central empirical premise behind global coherence analysis.
Springer →The 12-year cumulative PEAR review. Across more than 100 operators and 1,000 experimental series, the aggregate deviation from chance reached p < 0.0001. This is one of the strongest laboratory foundations for later GCP and CESM work.
Full text (PDF) →A rigorous meta-analysis of 832 human-REG studies across 68 laboratories, finding a small but consistent and statistically robust effect. It provides the broadest early empirical basis for treating mind-randomness interaction as a replicable statistical question.
DOI: 10.1007/BF00732509 →PEAR's first major peer-reviewed paper, establishing the experimental protocol and reporting aggregate results from thousands of operator-REG trials. It introduced the engineering approach to studying consciousness-related anomalies in random physical systems.
Full text (PDF) →The first clean experimental demonstration of human interaction with a quantum random process. Schmidt used a device based on radioactive decay and reported above-chance anticipation of random outputs, launching the modern electronic RNG research programme.
PA Archives →First, the platform-specific market results connecting GCP coherence measures to equity markets. Second, the broader finance literature showing that attention, sentiment, and narratives leave measurable market footprints.
These papers form the direct empirical basis for the market models on this platform, moving from the initial discovery of a GCP-stock return relationship to robustness checks, attention validation, and the peer-reviewed VIX/returns model foundation.
The primary market publication underlying the platform's forecast models. Demonstrates that GCP1 Max[Z] co-varies with VIX and S&P 500 returns at p < 0.01 across multiple consecutive days. The paper also reports out-of-sample trading simulations in which GCP-based portfolios outperform baseline strategies.
DOI: 10.1108/JES-11-2023-0663 →Replication and robustness check of the original market finding. Removes outliers, applies global stock market data, and confirms that Max[Z] remains significantly related to equity returns across the extended 1999–2020 sample.
Full text (PDF) →The founding empirical paper of the market research programme. First peer-reviewed demonstration that a variable derived from GCP random number data is statistically related to stock market index returns.
PhilPapers record →These papers provide the mainstream finance foundation for interpreting collective attention, sentiment, and narratives as market-relevant variables.
Argues that contagious narratives and collective attention are major drivers of macroeconomic fluctuations. This provides the broadest mainstream economic framing for treating collective cognition as a market force.
DOI: 10.1257/aer.107.4.967 →Develops the Opening Range Breakout framework used as the methodological precursor to the platform's intraday market-confirmation component. It provides the finance-side foundation for treating early-session price discovery as statistically informative.
ScienceDirect →Establishes Google search volume as a direct, real-time measure of investor attention, predicting short-term price pressure and subsequent reversals. This is the methodological bridge for treating GCP Max[Z] as an attention-adjacent market variable.
DOI: 10.1111/j.1540-6261.2011.01679.x →Shows that attention-grabbing events, news, abnormal volume, and extreme returns drive predictable investor buying behavior. Supports the broader idea that collective attention leaves measurable market traces.
DOI: 10.1093/rfs/hhm079 →Shows that financial media pessimism predicts next-day trading volume and short-term return reversals, grounding the platform's use of sentiment-adjacent and attention-adjacent variables in established empirical finance.
DOI: 10.1111/j.1540-6261.2007.01232.x →Develops a formal model of limited investor attention showing that when attention concentrates on market-wide categories, co-movements amplify. This provides the asset-pricing logic for why broad attention shocks could show up in cross-market dynamics.
DOI: 10.1016/j.jfineco.2005.05.003 →The out-of-sample evaluation underlying this platform is an ongoing pre-registered study on the Open Science Framework: DOI: 10.17605/OSF.IO/JK24C. Pre-registration locks the hypotheses, sample period, and analysis plan before data is observed. The study is currently accumulating out-of-sample observations — findings will be reported as the record grows.
Coherence Analytics is built around a simple but unusual empirical question: do random systems sometimes show structured deviations around moments of heightened human attention, shared emotional significance, or collective events? The platform does not assume the answer. It gives researchers and curious users tools to explore the question carefully.
The platform brings together event exploration, device-level RNG analysis, model comparison, and hypothesis drafting. It is designed to help users move from an observation to a clearer empirical question.
The outputs are exploratory research results. They are not proof of causality, not investment advice, and not claims that any single event caused a signal. The point is disciplined testing, not dramatic conclusions.
The site is for people who want to inspect the data themselves: start with one device, one event day, one model comparison, or one hypothesis. No commitment to the theory is required.
Coherence Analytics is motivated by peer-reviewed work on the Global Consciousness Project, RNG deviations, collective attention, market behaviour, and the Cognitive Entropy Shift Model. The platform treats these ideas as empirical questions. Patterns should be compared against baselines, checked across time, and interpreted with care.
Ulf Holmberg is a Swedish economist with a PhD in economics and a Master’s degree in Statistics. His work spans academic research, central banking, financial-sector development, macro-financial risk analysis, and independent research through Entropy Analytics.
His research programme explores whether coherence-related measures from random systems can be studied in relation to collective attention, financial markets, and other real-world systems. The work includes peer-reviewed publications in consciousness-related science and financial economics.
Coherence Analytics is for research, exploration, and hypothesis generation. It does not provide investment advice or claims of established causality.
For research enquiries: ulf.e.holmberg@me.com
The builder uses the latest data already loaded in this browser session. Choose the source that matches the question you want to formalize.
Use this when you want to test whether a selected Peak deviation day aligns with independently documented events. This builder intentionally excludes market-model results.
Use this when you want to formalize whether the GCP2-augmented model differs from the market-only baseline for the selected trading day.
Use this when you want to formalize a local device result: one device, one selected subset, and one comparison baseline.
