stylefact package

Submodules

stylefact.finance module

The toolkit for the statistical laws of financial time-series

stylefact.finance.autocorrelation(series, max_lag=1000, lags=None)

Autocorrelation function f(k) measures the pearson correlation of two variables with lag k.

Parameters:
  • series (array-like) – time-series to be evaluated
  • max_lag (int, optional) – maximum lag evaluated
Returns:

  • lags (list) – list of lags evaluated
  • acf_values (list) – list of correlation values

Examples

References

stylefact.finance.coarsefine_volatility(series, delta=5, lags=None, min_lag=-20, max_lag=20)

Coarse fine volatility

Parameters:
  • series (array-like) – time-series of price return
  • delta (int, optional) – the length to calculate coarse price return
  • min_lag (int, optional) – minimum lag to evaluate Default is -20
  • max_lag (int, optional) – maximum lag to evaluate Default is 20
Returns:

  • lags (list) – list of lags
  • values (list) – list of values corresponding to the lags

References

[Ulrich]Ulrich A. Müller et al., Volatilities of different time resolutions — Analyzing the dynamics of market components

Journal of Empirical Finance Volume 4, Issues 2–3, June 1997, Pages 213-239

Examples

stylefact.finance.gainloss_asymmetry(series, theta=0.1)

Gain loss asymmetry

Parameters:
  • series (array-like) – time-series to be evaluated
  • theta (int, optional) – The hyper-paramter theta
Returns:

  • step_dist_p (numpy array) – The distribution of required time to reach positive theta price change
  • step_dist_n (numpy array) – The distribution of required time to reach negative theta price change

References

[Mogens]Mogens H. Jensen et al., Inverse statistics in economics: The gain-loss asymmetry Physica A 324 (1) 338-343 2003.

Examples

stylefact.finance.leverage_effect(series, max_lag=50, lags=None)

The leverage effect, the lead-lag correlation between price return and volatility.

Parameters:
  • series (list) – list of symbols to be analyzed
  • max_lag (int, optional) – maximum lag evaluated
  • lags (list, optional) – lags to evaluate leverage effect
Returns:

  • lags (list) – list of ranks
  • lev_values (list) – values of leverage effect

Examples

References

stylefact.finance.linear_distribution(series)
Parameters:series (array-like) – list of symbols to be analyzed
Returns:
  • ticks (list) – list of ranks
  • dist_values (list) – frequency of i-th most frequent words

Examples

References

stylefact.finance.log_distribution(series, side='positive', ticks=None, sample_point=100)
Parameters:
  • series (array-like) – time-series to be evaluated
  • side (str (positive,negative), optional) – the side to evaluate the tail
  • ticks (array-like, optional) – hoge
  • sample_point (int, optional) – If ticks is not specified, the number of ticks is set to this value Default 100
Returns:

  • ticks (array-like)
  • dist_values (list) – the probability between ticks[i] and ticks[i+1]

Examples

References

stylefact.utils module

stylefact.utils.generate_x(max_x=1000, mode='log')
stylefact.utils.loglog_fitting(x, y)

stylefact.visualize module

stylefact.visualize.autocorrelation(x, y, filename, scale='log')
stylefact.visualize.coarsefine_volatility(x, y, filename)
stylefact.visualize.gainloss_asymmetry(y1, y2, filename, xscale='log')
stylefact.visualize.leverage_effect(x, y, filename)
stylefact.visualize.linear_distribution(x, y, filename)
stylefact.visualize.log_distribution(x, y, filename)

Module contents

Module level accessible objects.