stpnode properties

STP node iconThe Spatio-Temporal Prediction (STP) node uses data that contains location data, input fields for prediction (predictors), a time field, and a target field. Each location has numerous rows in the data that represent the values of each predictor at each time of measurement. After the data is analyzed, you can use it to predict target values at any location within the shape data that's used in the analysis.

Table 1. stpnode properties
stpnode properties Data type Property description
Fields properties    
target field This is the target field.
location field The location field for the model. Only geospatial fields are allowed.
location_label field The categorical field to be used in the output to label the locations chosen in location
time_field field The time field for the model. Only fields with continuous measurement are allowed, and the storage type must be time, date, timestamp, or integer.
inputs [field1 ... fieldN] A list of input fields.
Time Interval properties    
interval_type_timestamp Years Quarters Months Weeks Days Hours Minutes Seconds  
interval_type_date Years Quarters Months Weeks Days  
interval_type_time Hours Minutes Seconds Limits the number of days per week that are taken into account when creating the time index that STP uses for calculation
interval_type_integer Periods (Time index fields only, Integer storage) The interval to which the data set will be converted. The selection available is dependent on the storage type of the field that is chosen as the time_field for the model.
period_start integer  
start_month January February March April May June July August September October November December The month the model will start to index from (for example, if set to March but the first record in the data set is January, the model will skip the first two records and start indexing at March.
week_begins_on Sunday Monday Tuesday Wednesday Thursday Friday Saturday The starting point for the time index created by STP from the data
days_per_week integer Minimum 1, maximum 7, in increments of 1
hours_per_day integer The number of hours the model accounts for in a day. If this is set to 10, the model will start indexing at the day_begins_at time and continue indexing for 10 hours, then skip to the next value matching the day_begins_at value, etc.
day_begins_at 00:00 01:00 02:00 03:00 ... 23:00 Sets the hour value that the model starts indexing from.
interval_increment 1 2 3 4 5 6 10 12 15 20 30 This increment setting is for minutes or seconds. This determines where the model creates indexes from the data. So with an increment of 30 and interval type seconds, the model will create an index from the data every 30 seconds.
data_matches_interval Boolean If set to N, the conversion of the data to the regular interval_type occurs before the model is built. If your data is already in the correct format, and the interval_type and any associated settings match your data, set this to Y to prevent the conversion or aggregation of your data. Setting this to Y disables all of the Aggregation controls.
agg_range_default Sum Mean Min Max Median 1stQuartile 3rdQuartile This determines the default aggregation method used for continuous fields. Any continuous fields which are not specifically included in the custom aggregation will be aggregated using the method specified here.
custom_agg [[field, aggregation method],[]..] Demo: [['x5' 'FirstQuartile']['x4' 'Sum']] Structured property: Script parameter: custom_agg For example: set :stpnode.custom_agg = [ [field1 function] [field2 function] ] Where function is the aggregation function to be used with that field.
Basic properties    
include_intercept flag  
max_autoregressive_lag integer Minimum 1, maximum 5, in increments of 1. This is the number of previous records required for a prediction. So if set to 5, for example, then the previous 5 records are used to create a new forecast. The number of records specified here from the build data are incorporated into the model and, therefore, the user does not need to provide the data again when scoring the model.
estimation_method Parametric Nonparametric The method for modeling the spatial covariance matrix
parametric_model Gaussian Exponential PoweredExponential Order parameter for Parametric spatial covariance model
exponential_power number Power level for PoweredExponential model. Minimum 1, maximum 2.
Advanced properties    
max_missing_values integer The maximum percentage of records with missing values allowed in the model.
significance number The significance level for hypotheses testing in the model build. Specifies the significance value for all the tests in STP model estimation, including two Goodness of Fit tests, effect F-tests, and coefficient t-tests.
Output properties    
model_specifications flag  
temporal_summary flag  
location_summary flag Determines whether the Location Summary table is included in the model output.
model_quality flag  
test_mean_structure flag  
mean_structure_coefficients flag  
autoregressive_coefficients flag  
test_decay_space flag  
parametric_spatial_covariance flag  
correlations_heat_map flag  
correlations_map flag  
location_clusters flag  
similarity_threshold number The threshold at which output clusters are considered similar enough to be merged into a single cluster.
max_number_clusters integer The upper limit for the number of clusters which can be included in the model output.
Model Options properties    
use_model_name flag  
model_name string  
uncertainty_factor number Minimum 0, maximum 100. Determines the increase in uncertainty (error) applied to predictions in the future. It's the upper and lower bound for the predictions.