Marketplace›Web and tech›Data and analytics›Bimodal spectral analysis and anchor score report for time series
Bimodal spectral analysis and anchor score report for time series

Bimodal spectral analysis and anchor score report for time series

Data and analytics · delivered in 3-10 days

Basic - 100.99 USD, 3 days

  • 1 single-channel time series file reviewed
  • Anchor score (0-1)
  • Peak frequencies (f1, f2)
  • PDF analysis report
  • 1 round of follow-up questions
  • 1 revision

Pro - 500.99 USD, 5 days

  • Up to 10 files reviewed in one batch
  • Anchor score (0-1) per file
  • Peak frequencies (f1, f2) per file
  • Cross-file comparison table
  • PDF report with summary statistics
  • 2 rounds of follow-up questions
  • 2 revisions

Enterprise - 1875.99 USD, 10 days

  • 30+ files or a continuous time series
  • Anchor score (0-1) per file or per window
  • Custom frequency band configuration
  • Batch comparison and trend detection
  • Delivery as JSON, CSV or PDF
  • 3 rounds of follow-up questions
  • Optional monthly monitoring subscription
  • 3 revisions

You provide the domain the order is for and your brief.

Detecting hidden two-peak structure in a single channel

We examine your single-channel time series for bimodal spectral structure, meaning two frequency peaks with a valley between them, using a model-free method described in a published preprint (DOI: 10.5281/zenodo.22850972). Instead of raw code, you receive the finished analysis.

What you receive

  • An anchor score between 0 and 1 showing how strongly the bimodal pattern is present
  • Peak frequencies f1 and f2, expressed in Hz
  • A full PDF write-up of the analysis
  • Optional batch processing across several channels or files

What stays with us

The underlying algorithm, its exact parameter values and any tuning approach are kept as our own commercial asset; only the results are handed over.

How the method has been checked

  • On synthetic signals it correctly told apart single-peak, bimodal and noisy cases
  • Applied to LIGO O1 and O4c1 detector data, it produced no false positives on broadband noise
  • Using a 24-subject meditation EEG dataset (OpenNeuro ds001787), it found a significant expert-versus-novice group difference (p = 0.034), while 5 established measures, namely theta power, coefficient of variation, sample entropy, coherence and simple peak counting, all stayed above p = 0.2
  • Repeating the test across different parameters, frequency bands, window lengths and channel subsets showed the result held up

What we need from you

  • A single-column CSV file of your time series
  • Its sampling rate in Hz
  • The frequency band you are interested in, for example 3-8 Hz or 50-80 Hz

Typical turnaround

A single file is normally ready within 24 hours, while a batch of 10 or more files takes 3 to 5 business days; larger or ongoing work can extend to 10 days.

Where this can help

Applications range from EEG and neuroscience work such as meditation research, sleep staging, anaesthesia monitoring and neurological screening, through vibration and fault detection in rotating machinery, to regime-shift and cycle analysis in finance, and any other single-channel signal where two coexisting frequency components matter.

Indicative pricing

A single analysis starts at 80 USD, batch work from 600 USD, and ongoing collaboration or API integration is quoted on request.

Contact reference

ORCID 0009-0008-4540-1381 is listed for identification.

Scope of the service

This is a data analysis service only; it is not medical diagnosis, financial advice or an investment recommendation, and results are meant for research and informational use.

What happens after you buy

Your order opens its own thread here the moment it is paid, and everything about that order - questions, changes and the final report - happens in it.

Questions people ask

Do you provide the source code?

No, we do not share the algorithm. It is retained as our commercial asset, and its parameters are not disclosed. You receive the analysis output rather than the underlying code.

What data format do you accept?

We work from a single-column CSV of numeric values. Tell us the sampling rate in Hz and the frequency band you want examined. If your file is in another format such as Excel, txt or bdf, get in touch before ordering.

How long does it take?

A single file is typically completed within 3 days. Batches of 10 or more files take 3 to 5 business days. Enterprise work covering 30 or more files, or continuous data, can take up to 10 days. Faster delivery can be arranged as an add-on.

What makes this different from standard spectral analysis?

Conventional tools such as power spectral density, coefficient of variation, sample entropy or coherence measure overall power, stability or coupling, not whether a signal actually holds two separated peaks. Our anchor score targets that specific structure directly, on a 0 to 1 scale, and can be used alongside those other measures rather than instead of them.

Do you offer refunds?

We offer a refund if the analysis cannot be completed because of data format problems. Once the finished report has been delivered, refunds are not available, so please check with us first if you are uncertain about your data.

Can I use the results in a publication?

Yes, the results can be used in academic work. We ask that you reference the underlying method and its DOI, 10.5281/zenodo.22850972, in your citation.

People also buy

Data and analytics

Data analysis using Excel, SQL, and Power BI

We clean, analyse and visualise your business data using Excel, SQL and Power BI, delivering clear dashboards and actionable reports within 3 days.

$25.99 - $50.99PACKAGES FROM 3 days