tcmnode properties

TCM node iconTemporal causal modeling attempts to discover key causal relationships in time series data. In temporal causal modeling, you specify a set of target series and a set of candidate inputs to those targets. The procedure then builds an autoregressive time series model for each target and includes only those inputs that have the most significant causal relationship with the target.

Table 1. tcmnode properties
tcmnode Properties Values Property description
custom_fields Boolean  
dimensionlist [dimension1 ... dimensionN]  
data_struct
Multiple
Single
 
metric_fields fields  
both_target_and_input [f1 ... fN]  
targets [f1 ... fN]  
candidate_inputs [f1 ... fN]  
forced_inputs [f1 ... fN]  
use_timestamp
Timestamp
Period
 
input_interval
None
Unknown
Year
Quarter
Month
Week
Day
Hour
Hour_nonperiod
Minute
Minute_nonperiod
Second
Second_nonperiod
 
period_field string  
period_start_value integer  
num_days_per_week integer  
start_day_of_week Sunday
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
 
num_hours_per_day integer  
start_hour_of_day integer  
timestamp_increments integer  
cyclic_increments integer  
cyclic_periods list  
output_interval
None
Year
Quarter
Month
Week
Day
Hour
Minute
Second
 
is_same_interval
Same
Notsame
 
cross_hour Boolean  
aggregate_and_distribute list  
aggregate_default
Mean
Sum
Mode
Min
Max
 
distribute_default
Mean
Sum
 
group_default
Mean
Sum
Mode
Min
Max
 
missing_imput
Linear_interp
Series_mean
K_mean
K_meridian
Linear_trend
None
 
k_mean_param integer  
k_median_param integer  
missing_value_threshold integer  
conf_level integer  
max_num_predictor integer  
max_lag integer  
epsilon number  
threshold integer  
is_re_est Boolean  
num_targets integer  
percent_targets integer  
fields_display list  
series_dispaly list  
network_graph_for_target Boolean  
sign_level_for_target number  
fit_and_outlier_for_target Boolean  
sum_and_para_for_target Boolean  
impact_diag_for_target Boolean  
impact_diag_type_for_target
Effect
Cause
Both
 
impact_diag_level_for_target integer  
series_plot_for_target Boolean  
res_plot_for_target Boolean  
top_input_for_target Boolean  
forecast_table_for_target Boolean  
same_as_for_target Boolean  
network_graph_for_series Boolean  
sign_level_for_series number  
fit_and_outlier_for_series Boolean  
sum_and_para_for_series Boolean  
impact_diagram_for_series Boolean  
impact_diagram_type_for_series
Effect
Cause
Both
 
impact_diagram_level_for_series integer  
series_plot_for_series Boolean  
residual_plot_for_series Boolean  
forecast_table_for_series Boolean  
outlier_root_cause_analysis Boolean  
causal_levels integer  
outlier_table
Interactive
Pivot
Both
 
rmsp_error Boolean  
bic Boolean  
r_square Boolean  
outliers_over_time Boolean  
series_transormation Boolean  
use_estimation_period Boolean  
estimation_period
Times
Observation
 
observations list  
observations_type
Latest
Earliest
 
observations_num integer  
observations_exclude integer  
extend_records_into_future Boolean  
forecastperiods integer  
max_num_distinct_values integer  
display_targets
FIXEDNUMBER
PERCENTAGE
 
goodness_fit_measure
ROOTMEAN
BIC
RSQUARE
 
top_input_for_series Boolean  
aic Boolean  
rmse Boolean  
date_time_field field Time/Date field
auto_detect_lag Boolean This setting specifies the number of lag terms for each input in the model for each target.
numoflags Integer By default, the number of lag terms is automatically determined from the time interval that is used for the analysis.
re_estimate Boolean If you already generated a temporal causal model, select this option to reuse the criteria settings that are specified for that model, rather than building a new model.
display_targets
"FIXEDNUMBER"

"PERCENTAGE"
By default, output is displayed for the targets that are associated with the 10 best-fitting models, as determined by the R square value. You can specify a different fixed number of best-fitting models or you can specify a percentage of best-fitting models.