# Keep Macro Research Reproducible with OpenBB and FXMacroData

Canonical: https://fxmacrodata.com/articles/using-fxmacrodata-with-openbb-fx-macro-research
Author: FXMacroData
Published: 2026-06-22T11:00:11Z

Use the published OpenBB integration to preserve an identified macro research input while reviewing your workspace and reporting workflow.

**Quick answer:**Use the published FXMacroData Python package's OpenBB integration when you want macro observations inside an OpenBB research workflow. Keep the workspace choice separate from the identity and scope of the data.

A research workspace brings related tasks together, but it does not make all the data inside it equivalent. A policy-rate series, a market-price series and a saved research note each have a different role. When a team changes its workspace or reviews a provider connection, the useful question is which parts of the research process need to travel together.

OpenBB's announcement about its future makes that inventory timely. Its[published transition announcement](https://openbb.co/blog/openbb-belongs-to-everyone/) describes plans for the technology and says hosted customers will receive information directly. Use the provider's current notices for account-specific arrangements. This guide focuses on the practical data workflow, rather than assuming a particular hosting deadline or support outcome.

**Who this is for:**Analysts already using OpenBB who want to retain a traceable macro research input while reviewing or extending their setup.

## 1. Separate the workspace from the research input

Start with a saved report or notebook that the team actually uses. Identify the datasets it depends on, the series requested and the transformations that produce its conclusions. A migration is easier to review when the question is concrete: can the next setup reproduce this identified research input and keep its meaning intact?

Keep the dashboard layout, source observations and derived calculations as separate items in that inventory. A chart that still renders does not prove that it retained the same unit, date range or revision choice. Conversely, preserving a series does not automatically preserve every feature of a hosted workspace.

Preserve the research question while choosing the workspace
- Identify the report and its required series

- Choose the supported OpenBB environment

- Retrieve a bounded macro input

- Compare scope and interpretation

## 2. Add the available FXMacroData integration

The published`fxmacrodata`Python package includes an OpenBB provider and router. Use a Python environment compatible with the installed package and the OpenBB requirements. Its documented installation path is:

```
pip install "fxmacrodata[openbb]"
openbb-build
```

Install through your project's approved dependency workflow and record the resolved package versions. The command exposes the integration in your own supported OpenBB environment; it does not establish an OpenBB marketplace listing or supply another vendor's hosted account.

Supply a subscription credential through the environment or the OpenBB credential mechanism documented for your installation. Keep it out of shared notebooks and exported workspace layouts. Begin with an appropriately scoped[USD](/united-states) request when evaluating access and output.

## 3. Request one identified macro series

The following uses the provider's public Python interface for[US inflation](/united-states/inflation#api-docs) . The historical start date is an explicit research boundary and requires the relevant access:

```
from openbb import obb

inflation = obb.fxmacrodata.macro_indicators(
    currency="USD",
    indicator="inflation",
    start_date="2025-01-01",
    provider="fxmacrodata",
).to_df()
print(inflation.head())
```

Inspect the resulting columns and metadata before selecting a sample. Retain the reference periods and units alongside the values. Check whether the response covers the requested window and whether pagination or access notices affect the input. A data frame is a useful interface, but its shape alone does not define the economic series.

Use a separate request when the analysis needs pair-level market context. Define the base and quote currencies and the intended window first. Align the two inputs deliberately; a date common to both tables does not establish that a revised macro value was available at the historical trading decision.

Research item What to preserve Macro observation Series identity, unit, reference period and version context Market series Instrument or pair identity, price basis and sample window Derived comparison Selection rule, transformation and interpretation Workspace view Labels and source links that explain the displayed result

## 4. Compare the research inputs before the charts

When moving an existing report, compare row scope, missing values and transformations before comparing the appearance of its plots. A difference can arise because one request returns a different window, one calculation drops missing observations or one version uses a revised history. Those are sample-definition questions, not evidence that the market relationship changed.

Write the comparison as a short research note. State the series, requested period, selected observations and the calculation they feed. Explain any material difference in coverage or availability. The note should make sense to a colleague who sees only the finished report, without requiring access to the configuration history of the workspace.

## 5. Choose the interaction that fits the task

Keep OpenBB when its broader research environment is useful to the team. The FXMacroData package also documents Workspace-related options for supported installations. Use those instructions for the particular environment instead of assuming that an example Python command reproduces a complete hosted dashboard.

For a conversational macro question, a compatible assistant can use the[hosted MCP connection](/documentation/mcp-server) . For an application with a fixed series and report format, a direct SDK or REST request can be simpler. These choices concern how the analyst works; they do not change the need to identify the underlying series and preserve the evidence used in the argument.

## 6. Finish with a reproducible research handoff

A good handoff says which report was preserved, which data it consumes and how another analyst can check the result. Retain the resolved dependency versions and the retrieved research input. Test an unavailable response and a later rerun so the next person can distinguish a failed refresh from an actual change in the series.

The outcome is a workspace decision made around a defined research requirement. Whether the team keeps OpenBB, changes its hosting arrangement or uses a lighter macro workflow, the observation, its meaning and the calculation remain identifiable. That is the part of a research process worth making portable.

## Sources and further reading

- [OpenBB transition announcement](https://openbb.co/blog/openbb-belongs-to-everyone/)

- [Published FXMacroData package and OpenBB usage](https://pypi.org/project/fxmacrodata/)

- [OpenBB provider-extension documentation](https://docs.openbb.co/odp/python/extensions/providers)

AI was used for research, creating images and editing.

Source: FXMacroData