statsmodels.tsa.statespace.news.NewsResults¶
- class statsmodels.tsa.statespace.news.NewsResults(news_results, model, updated, previous, impacted_variable=None, tolerance=1e-10, row_labels=None)[source]¶
Impacts of data revisions and news on estimates of variables of interest
- Parameters:
- news_results
SimpleNamespace
instance
Results from KalmanSmoother.news.
- model
MLEResults
The results object associated with the model from which the NewsResults was generated.
- updated
MLEResults
The results object associated with the model containing the updated dataset.
- previous
MLEResults
The results object associated with the model containing the previous dataset.
- impacted_variable
str
,list
,array
,or
slice
,optional
Observation variable label or slice of labels specifying particular impacted variables to display in output. The impacted variable(s) describe the variables that were affected by the news. If you do not know the labels for the variables, check the endog_names attribute of the model instance.
- tolerance
float
,optional
The numerical threshold for determining zero impact. Default is that any impact less than 1e-10 is assumed to be zero.
- row_labelsiterable
Row labels (often dates) for the impacts of the revisions and news.
- news_results
References
[1]Bańbura, Marta, and Michele Modugno. “Maximum likelihood estimation of factor models on datasets with arbitrary pattern of missing data.” Journal of Applied Econometrics 29, no. 1 (2014): 133-160.
[2]Bańbura, Marta, Domenico Giannone, and Lucrezia Reichlin. “Nowcasting.” The Oxford Handbook of Economic Forecasting. July 8, 2011.
[3]Bańbura, Marta, Domenico Giannone, Michele Modugno, and Lucrezia Reichlin. “Now-casting and the real-time data flow.” In Handbook of economic forecasting, vol. 2, pp. 195-237. Elsevier, 2013.
- Attributes:
- total_impacts
pd.Series
Updates to forecasts of impacted variables from both news and data revisions, E[y^i | post] - E[y^i | previous].
- update_impacts
pd.Series
Updates to forecasts of impacted variables from the news, E[y^i | post] - E[y^i | revisions] where y^i are the impacted variables of interest.
- revision_impacts
pd.Series
Updates to forecasts of impacted variables from data revisions, E[y^i | revisions] - E[y^i | previous].
- news
pd.Series
The unexpected component of the updated data, E[y^u | post] - E[y^u | revisions] where y^u are the updated variables.
- weights
pd.Series
Weights describing the effect of news on variables of interest.
- revisions
pd.Series
The revisions betwen the current and previously observed data y^r_{revised} - y^r_{previous} where y^r are the revised variables.
- revision_weights
pd.Series
Weights describing the effect of revisions on variables of interest.
- update_forecasts
pd.Series
Forecasts based on the previous dataset of the variables that were updated, E[y^u | previous].
- update_realized
pd.Series
Actual observed data associated with the variables that were updated, y^u
- revised_prev
pd.Series
Previously observed data associated with the variables that were revised, y^r_{previous}
- revised
pd.Series
Currently observed data associated with the variables that were revised, y^r_{revised}
- prev_impacted_forecasts
pd.Series
Previous forecast of the variables of interest, E[y^i | previous].
- post_impacted_forecasts
pd.Series
Forecast of the variables of interest after taking into account both revisions and updates, E[y^i | post].
- revisions_iloc
pd.DataFrame
The integer locations of the data revisions in the dataset.
- revisions_ix
pd.DataFrame
The label-based locations of the data revisions in the dataset.
- updates_iloc
pd.DataFrame
The integer locations of the updated data points.
- updates_ix
pd.DataFrame
The label-based locations of updated data points.
- state_indexarray_like
Index of state variables used to compute impacts.
- total_impacts
Methods
get_details
([include_revisions, include_updates])get_impacts
([groupby, include_revisions, ...])summary
([impact_date, impacted_variable, ...])Create summary tables describing news and impacts
summary_details
([source, impact_date, ...])Create summary table with detailed impacts; by date, variable
summary_impacts
([impact_date, ...])Create summary table with detailed impacts from news; by date, variable
summary_news
([sparsify])Create summary table showing news from new data since previous results
summary_revisions
([sparsify])Create summary table showing revisions to the previous results' data
Properties
Revisions to data points that existed in the previous dataset
Updated data; new entries that did not exist in the previous dataset
Details of forecast revisions from news, organized by impacts first
Details of forecast revisions from news, organized by updates first
Impacts from news and revisions on all dates / variables of interest
Details of forecast revisions from revised data, organized by impacts
Details of forecast revisions from revisions, organized by updates