tseda.decomposition.ssa_result_summary module

SSA summary generation for diagnostics and observation logging text.

class tseda.decomposition.ssa_result_summary.SSAResultSummary(ssa_obj: Any, series: pandas.Series, window_size: int, eps: float = 1e-12)

Bases: object

Summarize rank-wise SSA explained/noise variance and AIC diagnostics.

Initialize the summary engine and compute rank-wise diagnostics.

Parameters:

• ssa_obj – Active SSA decomposition object.

• series – Input series used for baseline variance estimates.

• window_size – SSA window size (max rank cap).

• eps – Minimum positive floor used in log-variance terms.

__init__(ssa_obj: Any, series: pandas.Series, window_size: int, eps: float = 1e-12) → None

Initialize the summary engine and compute rank-wise diagnostics.

Parameters:

• ssa_obj – Active SSA decomposition object.

• series – Input series used for baseline variance estimates.

• window_size – SSA window size (max rank cap).

• eps – Minimum positive floor used in log-variance terms.

formulas() → dict[str, str]

Return symbolic formulas used in rank-based diagnostics.

Returns:

Mapping of short formula names to display strings.

plot_variance_explained() → plotly.graph_objects.Figure

Plot cumulative explained variance as a function of rank.

Returns:

Plotly figure for explained variance ratio vs rank.

plot_noise_variance() → plotly.graph_objects.Figure

Plot unexplained (noise) variance ratio as a function of rank.

Returns:

Plotly figure for noise variance ratio vs rank.

build_observation_text() → str

Compose the full auto-generated observation string from all section helpers.

Returns:

Multi-paragraph observation text suitable for the logging panel textarea.

Return type:

str