plot

plot

Opinionated static and interactive plots for quantitative research.

The first helpers focus on return streams: quick column charts, equity curves, drawdowns, and compact performance tearsheets. Directional label visualization and trade-log charts use the same interactive Plotly styling. Implementations live in :mod:qrt.plot.interactive and are re-exported here; this module only adds the ergonomic aliases :func:line, :func:performance, and :func:tearsheet. Return-stream statistics (performance, alpha/beta, rolling diagnostics, …) live in :mod:qrt.stats.

Functions

Name	Description
barchart	Plot one scalar value per Series item or selected DataFrame column.
correlation	Create an interactive scatterplot matrix for feature correlation analysis.
correlation_heatmap	Create an interactive correlation-matrix heatmap.
cumulative_returns	Create an interactive cumulative-returns chart, strategy vs an optional benchmark.
daily_returns	Create an interactive bar chart of periodic returns over time.
drawdown	Create an interactive underwater chart from periodic returns.
eoy_returns	Create an interactive bar chart of compounded end-of-year returns.
equity	Create an interactive compounded equity curve from periodic returns.
factor_contributions	Create an interactive stacked-area chart of cumulative factor return contributions.
factor_loadings	Create an interactive bar chart of full-period multi-factor regression coefficients.
labels	Plot directional labels over a price series.
line	Create an interactive Plotly line chart from selected numeric columns.
mae_mfe	Create an interactive MAE/MFE excursion scatter from a trade log.
metrics_table	Create an interactive table of performance metrics.
montecarlo	Create an interactive Monte Carlo fan chart of bootstrap-resampled return simulations.
montecarlo_distribution	Create interactive terminal-return and Max Drawdown distributions from a Monte Carlo run.
monthly_distribution	Create an interactive histogram of compounded monthly returns.
monthly_heatmap	Create an interactive, annotated calendar-month return heatmap.
noise_test	Create an interactive Noise Test fan chart of noise-perturbed return path simulations.
performance	Create an interactive Plotly equity-and-drawdown performance report.
precision_recall	Plot one-vs-rest multiclass precision-recall curves and averages.
psor	Plot the probability that Sortino exceeds a sweep of target thresholds.
psr	Plot the probability that Sharpe exceeds a sweep of target thresholds.
report	Create a full quantstats-style multi-panel strategy tearsheet.
return_quantiles	Create box plots comparing return distributions across compounding periods.
roc	Plot one-vs-rest multiclass ROC curves and micro/macro averages.
rolling_beta	Create an interactive chart of rolling beta over two windows (e.g. 6- and 12-month).
rolling_factor_betas	Create an interactive chart of rolling multi-factor betas over time.
rolling_sharpe	Create an interactive chart of annualized rolling Sharpe ratio.
rolling_sortino	Create an interactive chart of annualized rolling Sortino ratio.
rolling_volatility	Create an interactive chart of annualized rolling volatility.
show	Display a figure, optionally saving it to save_to as PNG (default) and/or self-contained HTML.
tearsheet	Alias for the interactive :func:performance report.
trade_distribution	Create a per-trade return distribution.
trades	Create an interactive price chart with entry/exit markers for a trade log.
tree_plot	Create historical and point-in-time performance treemaps for portfolio assets.
variance_test	Create an interactive Variance Testing fan chart of forward win-rate-varied trade paths.
worst_drawdowns	Create an equity curve with the worst drawdown periods shaded.

barchart

plot.barchart(
    data,
    columns=None,
    *,
    aggregate=None,
    sorted=False,
    color_scheme='sign',
    positive_color='#22C55E',
    negative_color='#EF4444',
    zero_color='#94A3B8',
    title=None,
    yaxis_title=None,
    label_angle=-90,
    height=500,
)

Plot one scalar value per Series item or selected DataFrame column.

A one-row DataFrame is plotted directly. For a DataFrame with multiple rows, aggregate must name a pandas reduction such as "sum", "mean", or "last", or be a callable applied to each column.

Parameters

Name	Type	Description	Default
data	pd.Series | pd.DataFrame	Numeric values as a Series, or observations in rows and categories in columns as a DataFrame.	required
columns	str | Iterable[str] | None	DataFrame column name(s) or shell-style pattern(s). Defaults to all numeric columns. Not used for Series input.	None
aggregate	str | Callable[[pd.Series], float] | None	Reduction applied to each selected DataFrame column when the frame has multiple rows. Required for multi-row frames.	None
sorted	bool	Whether to order bars from highest to lowest value.	False
color_scheme	str | Sequence[str] | Mapping[object, str]	"sign" for gain/loss colors, a single Plotly color, a sequence cycled across bars, or a category-to-color mapping.	'sign'
positive_color	str	Color for positive values in the "sign" scheme.	'#22C55E'
negative_color	str	Color for negative values in the "sign" scheme.	'#EF4444'
zero_color	str	Color for zero values in the "sign" scheme.	'#94A3B8'
title	str | None	Figure title. Defaults to the Series name or "Bar chart".	None
yaxis_title	str | None	Y-axis label.	None
label_angle	float	Category-label angle in degrees. Defaults to -90.	-90
height	int	Figure height in pixels.	500

Returns

Name	Type	Description
	Figure	A Plotly Figure.

correlation

plot.correlation(
    data,
    columns=None,
    *,
    color_by=None,
    color_discrete_sequence=None,
    color_discrete_map=None,
    color_continuous_scale='RdYlGn',
    hover_data=None,
    labels=None,
    triangle='full',
    diagonal=True,
    marker_size=7,
    marker_opacity=0.9,
    marker_line_width=0.6,
    marker_line_color='#1F2937',
    axis_color='#334155',
    axis_line_width=1.2,
    tick_length=5,
    tick_width=1.2,
    title='Feature relationships',
    height=None,
    width=None,
    panel_aspect_ratio=1.0,
)

Create an interactive scatterplot matrix for feature correlation analysis.

Linked selection highlights the same observations in every panel. color_by must name a column in data; QRT uses each row’s value in that column to determine the point color. Categorical values create discrete groups, while numeric values use a continuous scale. Numeric colors that span zero automatically center the scale at zero.

Parameters

Name	Type	Description	Default
data	pd.DataFrame	Observations in rows and features in columns.	required
columns	str | Iterable[str] | None	Numeric feature name(s) or shell-style pattern(s). By default, all numeric columns except color_by are included.	None
color_by	str | None	Exact DataFrame column name used to color observations. QRT does not infer a regime column or calculate category boundaries. String, categorical, and boolean values use discrete colors; numeric values use color_continuous_scale.	None
color_discrete_sequence	Sequence[str] | None	Colors assigned in order to categorical values. Defaults to the QRT categorical palette.	None
color_discrete_map	Mapping[object, str] | None	Exact value-to-color assignments for categories in the color_by column. Useful for stable semantic colors such as red for "risk-off".	None
color_continuous_scale	str | Sequence[str]	Plotly color-scale name or explicit color sequence for numeric values. Defaults to "RdYlGn".	'RdYlGn'
hover_data	Iterable[str] | None	Additional columns shown on hover, such as symbol or regime.	None
labels	dict[str, str] | None	Optional mapping from column names to display labels.	None
triangle	Literal['lower', 'upper', 'full']	Matrix panels to display: "lower", "upper", or "full".	'full'
diagonal	bool	Whether to show each feature plotted against itself.	True
marker_size	float	Marker diameter in pixels.	7
marker_opacity	float	Marker opacity from 0 (transparent) to 1 (opaque).	0.9
marker_line_width	float	Marker outline width in pixels.	0.6
marker_line_color	str	Marker outline color.	'#1F2937'
axis_color	str	Color used for subplot axes, ticks, and tick labels.	'#334155'
axis_line_width	float	Subplot axis-line width in pixels.	1.2
tick_length	float	Outward tick length in pixels.	5
tick_width	float	Tick width in pixels.	1.2
title	str	Figure title.	'Feature relationships'
height	int | None	Figure height in pixels. Inferred from the feature count by default.	None
width	int | None	Figure width in pixels. Overrides panel_aspect_ratio when set.	None
panel_aspect_ratio	float | None	Width-to-height ratio for each matrix panel. Defaults to 1.0 for square panels; use None for responsive container width.	1.0

Returns

Name	Type	Description
	Figure	A Plotly Figure.

correlation_heatmap

plot.correlation_heatmap(
    data,
    columns=None,
    *,
    method='pearson',
    min_periods=None,
    cluster=False,
    triangle='full',
    diagonal=True,
    color_continuous_scale='RdYlGn',
    show_values=None,
    value_format='.2f',
    text_size=11,
    labels=None,
    colorbar_label='Correlation',
    title='Feature correlation',
    cell_size=34,
    height=None,
    width=None,
)

Create an interactive correlation-matrix heatmap.

The color scale is pinned to [-1, 1] and centered at zero, so panels stay comparable across feature sets. Wide feature libraries are the main use case: cluster=True reorders rows and columns by similarity, which groups redundant features into visible blocks instead of scattering them.

Parameters

Name	Type	Description	Default
data	pd.DataFrame	Observations in rows and features in columns.	required
columns	str | Iterable[str] | None	Numeric feature name(s) or shell-style pattern(s). Defaults to all numeric columns.	None
method	Literal['pearson', 'spearman', 'kendall']	Correlation method: "pearson", "spearman", or "kendall". Rank methods resist outliers and capture monotonic relationships.	'pearson'
min_periods	int | None	Minimum overlapping observations required per pair.	None
cluster	bool	Whether to order features by hierarchical clustering so similar features are adjacent.	False
triangle	Literal['lower', 'upper', 'full']	Matrix cells to display: "lower", "upper", or "full".	'full'
diagonal	bool	Whether to show the self-correlation diagonal.	True
color_continuous_scale	str | Sequence[str]	Plotly color-scale name or explicit color sequence. Defaults to "RdYlGn".	'RdYlGn'
show_values	bool | None	Whether to annotate cells with correlation values. Defaults to annotating only matrices small enough to stay readable.	None
value_format	str	Format specification used for cell annotations.	'.2f'
text_size	float	Annotation font size in pixels.	11
labels	dict[str, str] | None	Optional mapping from column names to display labels.	None
colorbar_label	str	Color-bar title.	'Correlation'
title	str	Figure title.	'Feature correlation'
cell_size	float	Target cell size in pixels, used to infer the figure size.	34
height	int | None	Figure height in pixels. Inferred from the feature count by default.	None
width	int | None	Figure width in pixels. Inferred from the feature count by default.	None

Returns

Name	Type	Description
	Figure	A Plotly Figure.

cumulative_returns

plot.cumulative_returns(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    scale='linear',
    title=None,
    height=400,
)

Create an interactive cumulative-returns chart, strategy vs an optional benchmark.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
scale	Literal['linear', 'log', 'volatility_matched']	"linear" (plain cumulative percentage returns), "log" (log y-axis with percent tick labels, to compare compounding regimes across very different magnitudes), or "volatility_matched" (rescales the benchmark’s returns to match the strategy’s volatility before compounding, isolating differences in trend from differences in risk taken). The latter requires benchmark.	'linear'
title	str | None	Figure title.	None
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.

daily_returns

plot.daily_returns(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    active=False,
    title=None,
    height=320,
)

Create an interactive bar chart of periodic returns over time.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Benchmark periodic return series, aligned to returns on shared dates. Required when active=True.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
active	bool	Plot active returns (returns - benchmark) instead of raw returns. Requires benchmark.	False
title	str | None	Figure title.	None
height	int	Figure height in pixels.	320

Returns

Name	Type	Description
	Figure	A Plotly Figure.

drawdown

plot.drawdown(returns, *, return_type='simple', title='Drawdown', height=320)

Create an interactive underwater chart from periodic returns.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
title	str	Figure title.	'Drawdown'
height	int	Figure height in pixels.	320

Returns

Name	Type	Description
	Figure	A Plotly Figure.

eoy_returns

plot.eoy_returns(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    title=None,
    height=400,
)

Create an interactive bar chart of compounded end-of-year returns.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
title	str | None	Figure title.	None
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.

equity

plot.equity(
    returns,
    *,
    return_type='simple',
    title='Equity curve',
    label=None,
    height=450,
)

Create an interactive compounded equity curve from periodic returns.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
title	str	Figure title.	'Equity curve'
label	str | None	Series name for the equity curve. Defaults to returns.name.	None
height	int	Figure height in pixels.	450

Returns

Name	Type	Description
	Figure	A Plotly Figure.

factor_contributions

plot.factor_contributions(
    returns,
    factors,
    *,
    return_type='simple',
    rf='RF',
    covariance='nonrobust',
    title='Cumulative Factor Contribution',
    height=400,
)

Create an interactive stacked-area chart of cumulative factor return contributions.

See :func:qrt.stats.factor_contributions for the underlying per-period return attribution (alpha + one term per factor + residual, from a full-period fit). Cumulative sums are arithmetic (not compounded), so the stacked areas always add up exactly to the strategy’s cumulative excess return, drawn overlaid as a dotted reference line.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
factors	pd.DataFrame	DataFrame of periodic factor return columns, aligned to returns.	required
return_type	ReturnType	Whether returns is "simple" or "log" returns.	'simple'
rf	str | pd.Series | None	See :func:qrt.stats.factor_regression.	'RF'
covariance	CovarianceType	See :func:qrt.stats.factor_regression.	'nonrobust'
title	str	Figure title.	'Cumulative Factor Contribution'
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.

factor_loadings

plot.factor_loadings(
    returns,
    factors,
    *,
    return_type='simple',
    rf='RF',
    covariance='nonrobust',
    confidence_level=0.95,
    include_alpha=False,
    title='Factor Loadings',
    height=400,
)

Create an interactive bar chart of full-period multi-factor regression coefficients.

See :func:qrt.stats.factor_regression for the underlying calculation, e.g. the Fama-French five-factor model’s Mkt-RF/SMB/HML/RMW/CMA betas.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
factors	pd.DataFrame	DataFrame of periodic factor return columns, aligned to returns.	required
return_type	ReturnType	Whether returns is "simple" or "log" returns.	'simple'
rf	str | pd.Series | None	See :func:qrt.stats.factor_regression.	'RF'
covariance	CovarianceType	Standard-error estimator; see :func:qrt.stats.factor_regression.	'nonrobust'
confidence_level	float	Confidence level for the error bars. Defaults to 0.95.	0.95
include_alpha	bool	Whether to include the Alpha bar alongside the factor betas. Defaults to False since alpha’s scale/units usually differ from a factor beta (see :func:qrt.stats.factor_regression_stats for alpha shown on its own, annualized).	False
title	str	Figure title.	'Factor Loadings'
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.

labels

plot.labels(
    prices,
    labels,
    *,
    label_col='label',
    end_col=None,
    show_spans=True,
    title=None,
    height=520,
)

Plot directional labels over a price series.

Labels use green upward triangles for long (1), orange dashes for neutral (0), and red downward triangles for short (-1). When label data includes touch_time or end_time, translucent outcome spans can be drawn from each event to its endpoint.

Parameters

Name	Type	Description	Default
prices	pd.Series | pd.DataFrame	Close-price Series or frame containing a close column.	required
labels	pd.Series | pd.DataFrame	Event-indexed label Series or DataFrame containing label_col. Missing labels are omitted.	required
label_col	str	DataFrame column containing values in {-1, 0, 1}.	'label'
end_col	str | None	Optional outcome-end column. Defaults to touch_time when present, then end_time.	None
show_spans	bool	Whether to shade each event’s outcome interval when an endpoint column is available.	True
title	str | None	Figure title.	None
height	int	Figure height in pixels.	520

Returns

Name	Type	Description
	Figure	A Plotly Figure.

line

plot.line(data, columns=None, *, title=None, ylabel=None, height=450)

Create an interactive Plotly line chart from selected numeric columns.

Parameters

Name	Type	Description	Default
data	pd.Series | pd.DataFrame	Series or DataFrame to plot.	required
columns	str | Iterable[str] | None	Column name(s) or shell-style pattern(s) (e.g. "*_ret") to select. Defaults to all columns.	None
title	str | None	Figure title. Defaults to the column name (single column) or a generic title (multiple columns).	None
ylabel	str | None	Y-axis label.	None
height	int	Figure height in pixels.	450

Returns

Name	Type	Description
	Figure	A Plotly Figure.

mae_mfe

plot.mae_mfe(trades, *, title=None, height=430)

Create an interactive MAE/MFE excursion scatter from a trade log.

Two panels sharing the trade-return y-axis: max adverse excursion (how far each trade went against you – informs stop placement) and max favorable excursion (how much open profit existed – informs profit taking). Dashed identity lines mark the bounds return >= MAE and return <= MFE; exits near the MFE line captured most of their open profit, winners far below it round-tripped gains.

Parameters

Name	Type	Description	Default
trades	pd.DataFrame	Canonical trades-format DataFrame with mae/mfe columns.	required
title	str | None	Figure title.	None
height	int	Figure height in pixels.	430

Returns

Name	Type	Description
	Figure	A Plotly Figure.

metrics_table

plot.metrics_table(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    periods_per_year=None,
    rf=0.0,
    mode='full',
    title='Key Performance Metrics',
    height=None,
)

Create an interactive table of performance metrics.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates. Adds a benchmark column plus relative rows (Beta, Alpha, Correlation, R², Information Ratio, Tracking Error, Treynor Ratio) when given.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
rf	float	Annualized risk-free rate. Defaults to 0.0.	0.0
mode	Literal['basic', 'full']	"full" (default) for the complete quantstats-style metric set, "basic" for a compact headline subset. See :func:qrt.stats.metrics.	'full'
title	str	Figure title.	'Key Performance Metrics'
height	int | None	Figure height in pixels. Defaults to a size based on the number of metric rows.	None

Returns

Name	Type	Description
	Figure	A Plotly Figure.

montecarlo

plot.montecarlo(
    returns,
    sims=1000,
    *,
    return_type='simple',
    bust=None,
    goal=None,
    confidence=0.95,
    seed=None,
    block_size=None,
    periods=None,
    sample=200,
    title=None,
    height=520,
)

Create an interactive Monte Carlo fan chart of bootstrap-resampled return simulations.

Runs :func:qrt.stats.montecarlo on the full sims simulations, so the shaded confidence band and any bust/goal probabilities reflect the entire sample, then renders only a sample of the individual paths for readability. Unlike quantstats’ plot_montecarlo (which colors paths arbitrarily and, being built on a plain permutation of returns, can only vary each path’s shape, never its terminal value — see the note on :func:qrt.stats.montecarlo), each rendered path here is colored by its own outcome:

• Original (unresampled) path: bold black line, drawn on top.
• Busted paths (Max Drawdown breached bust): red.
• Reached goal paths (terminal return met goal): green.
• Everything else: light gray.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series (simple or log, per return_type).	required
sims	int	Number of simulated paths to run; statistics use all of them.	1000
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
bust	float | None	Optional Max Drawdown threshold (e.g. -0.2); breaching paths are colored red.	None
goal	float | None	Optional cumulative-return threshold (e.g. 1.0); paths reaching it are colored green and a reference line is drawn at that level.	None
confidence	float	Confidence level for the shaded fan. Defaults to 0.95.	0.95
seed	int | None	Optional random seed for reproducibility.	None
block_size	float | None	Optional mean block length (in periods) for a stationary block bootstrap, preserving autocorrelation/volatility clustering that i.i.d. resampling destroys. See the Note on :func:qrt.stats.montecarlo. Defaults to None (i.i.d.).	None
periods	int | None	Optional simulation horizon in periods, decoupled from len(returns) so a long, multi-regime history can be used as the resampling pool while each simulated path only covers a realistic forward horizon (e.g. 252 for one trading year). See the Note on :func:qrt.stats.montecarlo. Defaults to None (the full length of returns).	None
sample	int	Max number of individual simulated paths rendered (statistics still use all sims). Defaults to 200.	200
title	str | None	Figure title. Defaults to returns.name.	None
height	int	Figure height in pixels.	520

Returns

Name	Type	Description
	Figure	A Plotly Figure.

montecarlo_distribution

plot.montecarlo_distribution(
    returns,
    sims=1000,
    *,
    return_type='simple',
    bust=None,
    goal=None,
    confidence=0.95,
    seed=None,
    block_size=None,
    periods=None,
    title=None,
    height=420,
)

Create interactive terminal-return and Max Drawdown distributions from a Monte Carlo run.

Complements :func:montecarlo‘s fan chart with two histograms, both built from all sims simulations: the spread of terminal cumulative returns (reward) side-by-side with the spread of each simulation’s Max Drawdown (risk). Where quantstats’ plot_montecarlo_distribution shows only the former, pairing it with the drawdown distribution surfaces the downside risk a single terminal-value histogram hides. This distinction matters more than it first appears: since :func:qrt.stats.montecarlo resamples with replacement, the terminal-return panel here carries real information (a plain permutation-based simulation, as used by quantstats, cannot vary the terminal value at all — see the note on :func:qrt.stats.montecarlo). The original (unresampled) outcome is marked on both panels, along with the goal/bust thresholds and probabilities when given.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series (simple or log, per return_type).	required
sims	int	Number of simulated paths.	1000
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
bust	float | None	Optional Max Drawdown threshold (e.g. -0.2), marked on the drawdown panel.	None
goal	float | None	Optional cumulative-return threshold (e.g. 1.0), marked on the return panel.	None
confidence	float	Confidence level, forwarded to :func:qrt.stats.montecarlo for the reported probabilities. Defaults to 0.95.	0.95
seed	int | None	Optional random seed for reproducibility.	None
block_size	float | None	Optional mean block length (in periods) for a stationary block bootstrap, preserving autocorrelation/volatility clustering that i.i.d. resampling destroys. See the Note on :func:qrt.stats.montecarlo. Defaults to None (i.i.d.).	None
periods	int | None	Optional simulation horizon in periods, decoupled from len(returns) so a long, multi-regime history can be used as the resampling pool while each simulated path only covers a realistic forward horizon (e.g. 252 for one trading year). See the Note on :func:qrt.stats.montecarlo. Defaults to None (the full length of returns).	None
title	str | None	Figure title. Defaults to returns.name.	None
height	int	Figure height in pixels.	420

Returns

Name	Type	Description
	Figure	A Plotly Figure.

monthly_distribution

plot.monthly_distribution(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    title='Distribution of Monthly Returns',
    height=400,
)

Create an interactive histogram of compounded monthly returns.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates, overlaid as a second histogram.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
title	str	Figure title.	'Distribution of Monthly Returns'
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.

monthly_heatmap

plot.monthly_heatmap(
    returns,
    *,
    return_type='simple',
    title='Monthly returns',
    height=None,
)

Create an interactive, annotated calendar-month return heatmap.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series with a DatetimeIndex.	required
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
title	str	Figure title.	'Monthly returns'
height	int | None	Figure height in pixels. Defaults to a size based on the number of years.	None

Returns

Name	Type	Description
	Figure	A Plotly Figure.

noise_test

plot.noise_test(
    returns,
    sims=1000,
    *,
    return_type='simple',
    noise=0.1,
    bust=None,
    goal=None,
    confidence=0.95,
    seed=None,
    sample=200,
    title=None,
    height=520,
)

Create an interactive Noise Test fan chart of noise-perturbed return path simulations.

Runs :func:qrt.stats.noise_test on the full sims simulations, so the shaded confidence band and any bust/goal probabilities reflect the entire sample, then renders only a sample of the individual perturbed paths for readability. Unlike :func:montecarlo (which bootstrap-resamples returns, varying their order/composition) or :func:variance_test (which varies the win rate over a forward horizon), each rendered path here keeps the exact historical sequence of returns but jitters the magnitude of every period’s return by multiplicative noise, testing sensitivity to day-to-day noise/volatility rather than resampling variance:

• Real (original, unperturbed) path: bold black line, drawn on top.
• Busted paths (Max Drawdown breached bust): red.
• Reached goal paths (terminal return met goal): green.
• Everything else: light gray, labeled Random.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series (simple or log, per return_type).	required
sims	int	Number of simulated paths to run; statistics use all of them.	1000
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
noise	float	Standard deviation of the multiplicative noise applied to each period’s return. See the Args on :func:qrt.stats.noise_test.	0.1
bust	float | None	Optional Max Drawdown threshold (e.g. -0.2); breaching paths are colored red.	None
goal	float | None	Optional cumulative-return threshold (e.g. 1.0); paths reaching it are colored green and a reference line is drawn at that level.	None
confidence	float	Confidence level for the shaded fan. Defaults to 0.95.	0.95
seed	int | None	Optional random seed for reproducibility.	None
sample	int	Max number of individual simulated paths rendered (statistics still use all sims). Defaults to 200.	200
title	str | None	Figure title. Defaults to returns.name.	None
height	int	Figure height in pixels.	520

Returns

Name	Type	Description
	Figure	A Plotly Figure.

performance

plot.performance(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    periods_per_year=None,
    title=None,
    height=700,
)

Create an interactive Plotly equity-and-drawdown performance report.

Alias for :func:qrt.plot.interactive.performance.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
title	str | None	Figure title. Defaults to returns.name.	None
height	int	Figure height in pixels.	700

Returns

Name	Type	Description
	Figure	A Plotly Figure.

precision_recall

plot.precision_recall(
    y_true,
    y_score,
    *,
    classes=None,
    title='Multiclass precision-recall curve',
    height=500,
)

Plot one-vs-rest multiclass precision-recall curves and averages.

Parameters

Name	Type	Description	Default
y_true	Sequence[object] | pd.Series | np.ndarray	True class labels with shape (n_samples,).	required
y_score	pd.DataFrame | np.ndarray	Class scores with shape (n_samples, n_classes). DataFrame column names are used as class labels when classes is omitted.	required
classes	Iterable[object] | None	Class labels in the same order as the score columns.	None
title	str	Figure title.	'Multiclass precision-recall curve'
height	int	Figure height in pixels.	500

Returns

Name	Type	Description
	Figure	A Plotly Figure.

psor

plot.psor(
    returns,
    *,
    return_type='simple',
    periods_per_year=None,
    rf=0.0,
    threshold_max=None,
    points=121,
    title='Confidence That Sortino Exceeds Threshold',
    height=360,
)

Plot the probability that Sortino exceeds a sweep of target thresholds.

The probability uses only negative excess-return observations for its effective sample size and skewness correction.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
rf	float	Annualized risk-free rate and minimum acceptable return.	0.0
threshold_max	float | None	Upper annualized Sortino threshold. Defaults to at least 3.0, extending 25% beyond the observed Sortino when necessary.	None
points	int	Number of evenly spaced thresholds in the sweep.	121
title	str	Figure title.	'Confidence That Sortino Exceeds Threshold'
height	int	Figure height in pixels.	360

Returns

Name	Type	Description
	Figure	A Plotly Figure containing the PSoR threshold curve.

psr

plot.psr(
    returns,
    *,
    return_type='simple',
    periods_per_year=None,
    rf=0.0,
    threshold_max=None,
    points=121,
    title='Confidence That Sharpe Exceeds Threshold',
    height=360,
)

Plot the probability that Sharpe exceeds a sweep of target thresholds.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
rf	float	Annualized risk-free rate. Defaults to 0.0.	0.0
threshold_max	float | None	Upper annualized Sharpe threshold. Defaults to at least 3.0, extending 25% beyond the observed Sharpe when necessary.	None
points	int	Number of evenly spaced thresholds in the sweep.	121
title	str	Figure title.	'Confidence That Sharpe Exceeds Threshold'
height	int	Figure height in pixels.	360

Returns

Name	Type	Description
	Figure	A Plotly Figure containing the PSR threshold curve.

report

plot.report(
    returns,
    *,
    benchmark=None,
    return_type='simple',
    periods_per_year=None,
    rf=0.0,
    mode='full',
    worst_drawdowns_count=5,
    psr_threshold_max=None,
    psor_threshold_max=None,
    title=None,
)

Create a full quantstats-style multi-panel strategy tearsheet.

Lays out a chart column (cumulative-returns variants, EOY returns, return distributions, rolling risk diagnostics, drawdown analysis, a monthly heatmap) alongside a table column (key metrics, EOY returns, worst drawdowns) in a single scrollable Plotly figure. Individual panels are also available standalone (:func:metrics_table, :func:cumulative_returns, :func:eoy_returns, :func:monthly_distribution, :func:daily_returns, :func:rolling_volatility, :func:rolling_sharpe, :func:psr, :func:rolling_sortino, :func:psor, :func:rolling_beta, :func:worst_drawdowns, :func:drawdown, :func:monthly_heatmap, :func:return_quantiles).

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates. Adds benchmark overlays/columns and benchmark-relative panels (volatility-matched returns, rolling beta) throughout.	None
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
rf	float	Annualized risk-free rate. Defaults to 0.0.	0.0
mode	Literal['basic', 'full']	"full" (default) for the complete quantstats-style metrics table, "basic" for a compact headline subset. See :func:qrt.stats.metrics.	'full'
worst_drawdowns_count	int	Number of worst drawdown episodes to shade/list.	5
psr_threshold_max	float | None	Upper bound of the annualized Sharpe threshold sweep. Defaults to at least 3.0, extending 25% beyond the observed Sharpe ratio when necessary.	None
psor_threshold_max	float | None	Upper bound of the annualized Sortino threshold sweep. Defaults to at least 3.0, extending 25% beyond the observed Sortino ratio when necessary.	None
title	str | None	Figure title.	None

Returns

Name	Type	Description
	Figure	A single tall Plotly Figure stacking every tearsheet panel.

return_quantiles

plot.return_quantiles(
    returns,
    *,
    return_type='simple',
    title='Return Quantiles',
    height=400,
)

Create box plots comparing return distributions across compounding periods.

See :func:qrt.stats.distribution for the underlying period buckets.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
title	str	Figure title.	'Return Quantiles'
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.

roc

plot.roc(
    y_true,
    y_score,
    *,
    classes=None,
    title='Multiclass ROC curve',
    height=500,
)

Plot one-vs-rest multiclass ROC curves and micro/macro averages.

Parameters

Name	Type	Description	Default
y_true	Sequence[object] | pd.Series | np.ndarray	True class labels with shape (n_samples,).	required
y_score	pd.DataFrame | np.ndarray	Class scores with shape (n_samples, n_classes). DataFrame column names are used as class labels when classes is omitted.	required
classes	Iterable[object] | None	Class labels in the same order as the score columns.	None
title	str	Figure title.	'Multiclass ROC curve'
height	int	Figure height in pixels.	500

Returns

Name	Type	Description
	Figure	A Plotly Figure.

rolling_beta

plot.rolling_beta(
    returns,
    benchmark,
    *,
    return_type='simple',
    windows=(126, 252),
    title='Rolling Beta to Benchmark',
    height=320,
)

Create an interactive chart of rolling beta over two windows (e.g. 6- and 12-month).

See :func:qrt.stats.rolling_beta for the underlying calculation.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series	Benchmark periodic return series, aligned to returns on shared dates.	required
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
windows	tuple[int, int]	Pair of rolling window sizes, in periods (shorter first).	(126, 252)
title	str	Figure title.	'Rolling Beta to Benchmark'
height	int	Figure height in pixels.	320

Returns

Name	Type	Description
	Figure	A Plotly Figure.

rolling_factor_betas

plot.rolling_factor_betas(
    returns,
    factors,
    window=63,
    *,
    return_type='simple',
    rf='RF',
    min_observations=None,
    title=None,
    height=380,
)

Create an interactive chart of rolling multi-factor betas over time.

See :func:qrt.stats.rolling_factor_regression for the underlying calculation, e.g. the rolling Fama-French five-factor betas that reveal time-varying exposure.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
factors	pd.DataFrame	DataFrame of periodic factor return columns, aligned to returns.	required
window	int	Trailing window size, in periods. Defaults to 63 (~3 trading months).	63
return_type	ReturnType	Whether returns is "simple" or "log" returns.	'simple'
rf	str | pd.Series | None	See :func:qrt.stats.factor_regression.	'RF'
min_observations	int | None	See :func:qrt.stats.rolling_factor_regression.	None
title	str | None	Figure title. Defaults to a title naming window.	None
height	int	Figure height in pixels.	380

Returns

Name	Type	Description
	Figure	A Plotly Figure. Each factor’s hover also shows that window’s
	Figure	observation count and R², to distinguish real exposure changes from
	Figure	noisy/short-window estimation.

rolling_sharpe

plot.rolling_sharpe(
    returns,
    *,
    window=126,
    return_type='simple',
    periods_per_year=None,
    rf=0.0,
    title='Rolling Sharpe (6-Months)',
    height=320,
)

Create an interactive chart of annualized rolling Sharpe ratio.

See :func:qrt.stats.rolling_sharpe for the underlying calculation.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
window	int	Rolling window size, in periods.	126
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
rf	float	Annualized risk-free rate. Defaults to 0.0.	0.0
title	str	Figure title.	'Rolling Sharpe (6-Months)'
height	int	Figure height in pixels.	320

Returns

Name	Type	Description
	Figure	A Plotly Figure.

rolling_sortino

plot.rolling_sortino(
    returns,
    *,
    window=126,
    return_type='simple',
    periods_per_year=None,
    rf=0.0,
    title='Rolling Sortino (6-Months)',
    height=320,
)

Create an interactive chart of annualized rolling Sortino ratio.

See :func:qrt.stats.rolling_sortino for the underlying calculation.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
window	int	Rolling window size, in periods.	126
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
rf	float	Annualized risk-free rate. Defaults to 0.0.	0.0
title	str	Figure title.	'Rolling Sortino (6-Months)'
height	int	Figure height in pixels.	320

Returns

Name	Type	Description
	Figure	A Plotly Figure.

rolling_volatility

plot.rolling_volatility(
    returns,
    *,
    benchmark=None,
    window=126,
    return_type='simple',
    periods_per_year=None,
    title='Rolling Volatility (6-Months)',
    height=320,
)

Create an interactive chart of annualized rolling volatility.

See :func:qrt.stats.rolling_volatility for the underlying calculation.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
benchmark	pd.Series | None	Optional benchmark periodic return series, aligned to returns on shared dates, overlaid as a second line.	None
window	int	Rolling window size, in periods.	126
return_type	ReturnType	Whether returns/benchmark are "simple" or "log" returns.	'simple'
periods_per_year	int | None	Annualization frequency. Inferred from the index when not given.	None
title	str	Figure title.	'Rolling Volatility (6-Months)'
height	int	Figure height in pixels.	320

Returns

Name	Type	Description
	Figure	A Plotly Figure.

show

plot.show(
    figure,
    name=None,
    *,
    save_to=None,
    formats=('png',),
    width=1400,
    height=800,
    scale=2,
)

Display a figure, optionally saving it to save_to as PNG (default) and/or self-contained HTML.

Parameters

Name	Type	Description	Default
figure	Figure	Plotly figure to display (and optionally save).	required
name	str | None	File stem used when saving. Required if save_to is given.	None
save_to	str | Path | None	Directory to save the figure into. If None, the figure is only displayed.	None
formats	Iterable[str]	Output formats to save, any of "png" and/or "html".	('png',)
width	int	PNG width in pixels.	1400
height	int	PNG height in pixels.	800
scale	int	PNG scale factor (for higher-resolution exports).	2

tearsheet

plot.tearsheet(returns, **kwargs)

Alias for the interactive :func:performance report.

Parameters

Name	Type	Description	Default
returns	pd.Series	Strategy periodic return series.	required
**kwargs	Any	Forwarded to :func:performance (e.g. benchmark, title).	{}

Returns

Name	Type	Description
	Figure	A Plotly Figure.

trade_distribution

plot.trade_distribution(trades, by='exit_reason', *, title=None, height=430)

Create a per-trade return distribution.

With by=None, renders a sign-colored histogram with zero and mean reference lines. With a column name, renders the existing grouped box plot.

Parameters

Name	Type	Description	Default
trades	pd.DataFrame	Canonical trades-format DataFrame (see :mod:qrt.data.datasets).	required
by	str | None	Trades column to group by (e.g. "exit_reason", "direction", or any feature-snapshot column). Set to None for an overall returns-per-trade histogram.	'exit_reason'
title	str | None	Figure title.	None
height	int	Figure height in pixels.	430

Returns

Name	Type	Description
	Figure	A Plotly Figure.

Raises

Name	Type	Description
	ValueError	If by is not a trades column.

trades

plot.trades(trades, prices, *, features=None, title=None, height=520)

Create an interactive price chart with entry/exit markers for a trade log.

Renders the price series with one marker per trade entry (triangle up for longs, down for shorts) and exit (x, green for winners, red for losers), plus a translucent span shading each trade by outcome. Hovering a marker shows the trade’s reason, return, holding period, and any feature-snapshot columns beyond the reserved trades-format set.

Parameters

Name	Type	Description	Default
trades	pd.DataFrame	Canonical trades-format DataFrame (see :mod:qrt.data.datasets).	required
prices	pd.Series | pd.DataFrame	Close series (or OHLCV frame with a close column) with a DatetimeIndex covering the trade span.	required
features	pd.DataFrame | None	Optional datetime-indexed frame of indicator series (e.g. moving averages computed via :mod:qrt.indicator) drawn as overlay lines on the price axis.	None
title	str | None	Figure title.	None
height	int	Figure height in pixels.	520

Returns

Name	Type	Description
	Figure	A Plotly Figure.

tree_plot

plot.tree_plot(
    asset_returns,
    *,
    weights=None,
    title='Portfolio Performance Treemap',
    height=600,
)

Create historical and point-in-time performance treemaps for portfolio assets.

The default accumulated mode includes every asset held on or before each timestamp. Returns compound only while held, then remain at their terminal value after exit. Point-in-time mode shows only the assets held at the selected timestamp and their accumulated held-period returns through that date. A finite return marks a held period when weights are omitted; with weights, a finite return and nonzero finite weight are both required. Tiles have equal area unless weights are supplied. Dynamic weights use cumulative average absolute exposure in accumulated mode and current absolute exposure in point-in-time mode.

Parameters

Name	Type	Description	Default
asset_returns	pd.DataFrame	Wide frame of simple returns, indexed by timestamp with one asset per column.	required
weights	pd.DataFrame | pd.Series | None	Optional matching weight frame, or static weight series indexed by asset. Dynamic absolute weights are averaged cumulatively for tile area; static absolute weights are used directly.	None
title	str	Figure title.	'Portfolio Performance Treemap'
height	int	Figure height in pixels.	600

Returns

Name	Type	Description
	Figure	A Plotly Figure with a timestamp slider.

variance_test

plot.variance_test(
    trades,
    periods,
    sims=1000,
    *,
    return_type='simple',
    win_rate_variance=0.1,
    ruin=-0.9,
    confidence=0.95,
    seed=None,
    sample=200,
    title=None,
    height=520,
)

Create an interactive Variance Testing fan chart of forward win-rate-varied trade paths.

Runs :func:qrt.stats.variance_test on the full sims simulations, so the shaded confidence band and the make-money/ruin probabilities reflect the entire sample, then renders only a sample of the individual simulated paths for readability, alongside the actual historical ("Real") cumulative return over the most recent periods trades of trades for comparison (truncated to periods rather than the full history so its scale and length stay comparable to the simulated paths):

• Real (actual historical) path: bold black line, drawn on top.
• Ruined paths (Max Drawdown breached ruin): red.
• Profitable paths (positive terminal return): green.
• Everything else: light gray.

The simulated paths and confidence band are 0-based when computed (statistics such as make_money_probability/ruin_probability stay purely forward-looking, unaffected by history), but are rebased for display onto the last "Real" cumulative return so the fan chart continues smoothly from history instead of visually resetting to 0% at “Now” (the ruin reference line is shifted the same way).

When trades has a DatetimeIndex, the x-axis shows real calendar dates — the actual dates for "Real" and dates extrapolated forward from the last one for the simulated paths (marked with a dotted “Now” line at the boundary) — matching the date axes used throughout :mod:qrt.plot. Falls back to a plain trade-count axis otherwise.

Parameters

Name	Type	Description	Default
trades	pd.Series	Per-trade return series (simple or log, per return_type).	required
periods	int	Number of future trades to simulate per path.	required
sims	int	Number of simulated paths to run; statistics use all of them.	1000
return_type	ReturnType	Whether trades are "simple" or "log" returns.	'simple'
win_rate_variance	float	Amount by which each simulation’s win rate is randomly perturbed from the historical win rate. See the Args on :func:qrt.stats.variance_test.	0.1
ruin	float	Max Drawdown threshold (e.g. -0.9); breaching paths are colored red and marked with a reference line. See the Args on :func:qrt.stats.variance_test.	-0.9
confidence	float	Confidence level for the shaded fan. Defaults to 0.95.	0.95
seed	int | None	Optional random seed for reproducibility.	None
sample	int	Max number of individual simulated paths rendered (statistics still use all sims). Defaults to 200.	200
title	str | None	Figure title. Defaults to trades.name.	None
height	int	Figure height in pixels.	520

Returns

Name	Type	Description
	Figure	A Plotly Figure.

worst_drawdowns

plot.worst_drawdowns(
    returns,
    *,
    top=5,
    return_type='simple',
    title=None,
    height=400,
)

Create an equity curve with the worst drawdown periods shaded.

See :func:qrt.stats.drawdown_details for the underlying episode breakdown.

Parameters

Name	Type	Description	Default
returns	pd.Series	Periodic return series.	required
top	int	Number of worst (deepest) drawdown episodes to shade.	5
return_type	ReturnType	Whether returns are "simple" or "log" returns.	'simple'
title	str | None	Figure title.	None
height	int	Figure height in pixels.	400

Returns

Name	Type	Description
	Figure	A Plotly Figure.