marketflows.analysis.metrics

Metric calculations for analysis.

Computes normalized series (by base asset and first valid record), optional EMAs, derivatives (growth/inflection), optional smoothing, and optional unit scaling.

Functions

calculate_group_metrics(*, base_assets[, ...])

Calculate metrics for groups of or single assets.

calculate_range_metrics(*[, df_master, df_long])

Calculate metrics for market cap ranges.
marketflows.analysis.metrics.calculate_group_metrics(*, base_assets, df_base=None, df_groups, analysis_config)

Calculate metrics for groups of or single assets.

Given a dataframe of assets or groups of assets, normalize with base currency, normalize with time, calculate EMAs, calculate derivatives, and smooth with EMA (10 by default). The original data is normalized with the base assets. The derivatives are not normalized with the base assets.

Parameters:

  • base_assets (list[str]) – list of assets that will be used as base currency (us-dollar is default since all data points are in terms of USD)

  • df_groups (DataFrame) – dataframe with the groups or individual assets for which we are calculating metrics

  • df_base (DataFrame | None, default: None) – dataframe with all base token data

  • analysis_config (AnalysisConfig) – configuration parameters for analysis module

Return type:

DataFrame

Returns:

dataframe with all columns necessary for plotting

Raises:

  • ValueError – if base_assets are not in df_base

  • TypeError – if diff_orders are not all integers

marketflows.analysis.metrics.calculate_range_metrics(*, df_master=None, df_ranges, df_long=None, provider_config, analysis_config)

Calculate metrics for market cap ranges.

Given a dataframe of market cap ranges (each range is specified by its lower limit e.g. df_ranges.columns = [1e9, 1e10, 1e11]), normalize with base currency, normalize with time, calculate up to 2nd derivative, and smooth with EMAs. The data (original, 1st derivative, or 2nd derivative) is normalized with the original data for each base asset.

Parameters:

  • df_master (DataFrame | None, default: None) – dataframe with all individual asset data

  • df_ranges (DataFrame) – dataframe with the market cap ranges for which we are calculating metrics

  • df_long (DataFrame | None, default: None) – market caps for all the assets with one master datetime index organized into blocks of one asset after another

  • provider_config (ProviderConfig) – configuration parameters for provider module

  • analysis_config (AnalysisConfig) – configuration parameters for analysis module

Return type:

DataFrame

Returns:

dataframe with all columns necessary for plotting

Notes

  • If the first valid time is not a valid time for a column (this can happen when the valid records do not coincide for the base asset and the asset, causing a mostly missing record column), the first valid time for that column is used instead of the first valid time for all columns