Skip to main content
track_propagations This function analyses the propagation of directional changes across different timeframes to model the casual evolution.

​
Function Parameters

ParameterTypeRequiredDescription
df_listobjectTrueA list of dataframes including ‘datetime’ and ‘trend_identified’ columns ordered from shortest to longest to track propagations across multiple timeframes.
propagation_levelintTrueSet the propagation level to track for analysis e.g. propagation_level = 2 will reveal only the directional changes which propagate at least twice from the initial indicator.

​
Python Code Example

from sumtyme import EIPClient

# Replace 'your-subdomain' with your assigned EIP subdomain.
client = EIPClient(subdomain='your-subdomain',apikey_path='config.txt')

# Download sample API output data from Github
data_1m_output = pd.read_csv('https://raw.githubusercontent.com/sumteam/data_store/main/api_output/btcusdt_1m_proactive_outputs.csv') # 1 minute time series
data_2m_output = pd.read_csv('https://raw.githubusercontent.com/sumteam/data_store/main/api_output/btcusdt_2m_proactive_outputs.csv') # 2 minute time series
data_3m_output = pd.read_csv('https://raw.githubusercontent.com/sumteam/data_store/main/api_output/btcusdt_3m_proactive_outputs.csv') # 3 minute time series

# List of timeframes to analyse in order of timeframe: 1m --> 3m 
time_series_to_track = [data_1m_output, data_2m_output, data_3m_output]

# Track directional changes which propagate at least twice (from 1m to 3m)
latest_change = client.track_propagations(
    df_list= time_series_to_track
    propagation_level=2
)

# Returns a dataframe with columns: [datetime, trend_identified, propagation_level]