JPMaQS Cheat Sheet

M ACROSYNERGY

K EY DEFINITIONS K EY SUPPORT WEBSITES

What is JPMaQS JPMaQS Site:
JPMaQS stands for J.P. Morgan Macrosynergy Quantamental System. It is a service that makes it easy to use The official JPMaQS Site on J.P. Morgan Markets (requires password)
quantitative-fundamental (“quantamental”) information for trading, be it algorithmic strategies or discretio- https://markets.jpmorgan.com/#jpmaqs
nary trader support. It allows quick and cost-efficient backtesting. Quantamental Academy (free access)
Macro quantamental indicators https://macrosynergy.com/academy
Macro quantamental indicators are time series of macroeconomic information states designed for the deve- Understanding of quantamental indicators:
lopment and backtesting of financial markets trading strategies. https://macrosynergy.com/academy/what-are-macro-quantamental-indicators
Information state https://macrosynergy.com/academy/introductory-tutorials
Information state is the latest instance of a quantamental indicator based on its concurrent vintage assigned Quantamental indicators:
to the date at which it was available to the market. https://macrosynergy.com/academy/quantamental-indicators-on-jpmaqs
Vintage
Examples of trading strategies and use of data science:
A vintage is an instance of a full economic time series; economic history is recorded in vintages, i.e., time
https://macrosynergy.com/academy/examples-macro-trading-factors
series of time series. Vintages allow replicating what markets knew at any day in recent history, which is
https://macrosynergy.com/academy/statistics-packages-with-quantamental-indicators
critical for backtesting algorithmic strategies. Disregarding vintages leads to survivorship and look-ahead
biases in evaluating trading ideas. Kaggle - free limited dataset with notebooks:
Macro quantamental categories www.kaggle.com/datasets/macrosynergy/fixed-income-returns-and-macro-trends
Macro quantamental categories are sets of comparable quantamental indicators across multiple countries,
currency areas, or markets, also called data panels. A CCESS TO JPM A QS DATA
Themes
JPMaQS data are downloaded through J.P.Morgan DataQuery. APIs are available from J.P. Morgan and the
The theme is a broad set of category groups designed to facilitate browsing and discovery of categories for Macrosynergy package. Free trials are available. Data excluding the last few months are available free for
strategy research projects research. Please contact J.P. Morgan sales for trial ID and password. For download guide please visit
https://docs.macrosynergy.com
https://macrosynergy.com/academy/quantamental-indicators-on-jpmaqs

Economic trends daily categories designed to capture actual fundamental developments and to ATTRIBUTES OF DATA
de-emphasize data volatility and distortions
real date The date of the information state as observed by the markets
Macroeconomic balance categories of changes in economic conditions designed to capture actual fun-
value Value of the macro quantamental indicator
sheets damental development as opposed to data volatility
grading Grading of indicator in terms of point-in-time information content
Financial conditions categories that track the conditions of the broader financial system and its
eop_lag Days passed since the last date of the observation period for the underlying data
impact on the economy
mop_lag Days passed since the median date of the observation period for the underlying data
Shock and risk measures categories of changes in expectations, uncertainty, and risk aversion.
Stylized trading factors generic categories of basic trading strategy ideas based on macro quantamen-
tal indicators and conventional trading factors. JPM A QS TICKERS
Generic returns approximate daily profit and loss series of stylized derivatives positions in
percent of notional or risk capital. Every timeseries has a unique ticker with the same structure:
USD_INTRGDPv5Y_NSA_P1M1ML12_3MMA
Category Group
Market USD currency area or market
Within each theme, JPMaQS macro quantamental indicators are organised across multiple category groups
Base category INTRGDP basic concept, for example GDP growth
(e.g. Theme: Economic trends, Category: Consistent core CPI trends). Notebooks available for each on J.P.
Adjustment _NSA seasonal and other adjustment, else NSA
Morgan Markets
Trend v5Y reference to past aggregate, for example versys 5-year average
Change P1M1ML12 type and paramer of change, for example % last month over a
C ONTACTS J.P. M ORGAN C ONTACTS M ACROSYNERGY previous month 12 months ago
Filters 3MMA averages medians etc., for example 3-month moving average
[email protected] [email protected]
 JPMaQS Cheat Sheet
M ACROSYNERGY

M ACROSYNERGY PACKAGE M ACRO INVESTMENT INSPIRATIONS

https://github.com/macrosynergy/macrosynergy, docs.macrosynergy.com Macro trading signals collected here: https://macrosynergy.com/academy/examples-macro-trading-factors
Download - download required JPMaQS time series Every notebook has the same structure. Here is the example of Inflation as equity trading signal
JPMaQSDownload facilitates JPMaQS indicator download Summary and main ideas
Management - data availability - start years and missing info Adding inflation sensitivity to a hypothetical long-only portfolio resulted in reduced or fully avoided draw-
downs during the most turbulent periods and bolstered post-crisis performance without significantly redu-
check_availability visualizes start years and the number of missing values cing performance during normal periods.
missing_in_df displays missing categories and cross-sections
Get packages and JPMaQS data
update_df concatenates two JPMaQS data frames
Panel - new time series calculations, correlations, and visualizations imports standard Python packages and macrosynergy package
currency lists define required currencies
view_ranges overview of long-term series distributions in a panel
categories lists required lists of categories, such as inflation, and returns
view_timelines displays a facet grid of timeline charts of one or more categories
download code using client_id and client_secret and download module from macrosynergy package
view_heatmap_grades displays a colored table of grading quality of indicators
access JPMaQS and download earlier defined categories for earlier defined currencies
make_blacklist creates a standardized dictionary of blacklist periods, i.e., periods that affect the
availability check data for gaps, limitations, missing data, etc.
validity of indicators. This list can be passed to macrosynergy package functions
make_relative_value generates a data frame of relative values for a given list of categories. "relative” Transformations and checks
means that the original value is compared to a basket average simple calculations and transformations to derive relevant signals and targets, such as creating difference between cate-
panel_calculator simplifies applying transformations to each panel cross-section using a string- gories, z-scores, relative values etc.
based formula Features Create and display excess inflation, relative inflation, zn-scores
make_zn_scores normalizes values across different categories Targets define target - either directional or relative
linear_composite calculate linear combinations of different categories Value checks
categoryRelations visualization and analysis of two categories, two time-series panels crucial part of any notebook: evaluates performance of suggested strategies, based on economic theory, formulated in
correl_matrix visualizes two types of Pearson correlations: within category across cross- summary
sections and across categories Correlation and accuracy - check for indicator/ subsequent target and signal accuracy
historic_vol estimate historic annualized standard deviations of asset returns
Signal - accuracy and the strength of selected signals calculations
SignalReturnRelations analyze, visualize, and compare the relationships between panels of trading sig-
nals and panels of subsequent returns
signals_table produces table on relations of various signals with the target return
accuracy_bars plot bar chart for the overall and balanced accuracy metrics
correlation_bars plot correlation coefficients and significance
multiple_relations_table statistics for each return and signal category specified with each frequency and
aggregation method
Learning - machine learning solutions subpackage
panel_time_series_split produce, visualize and use walk-forward validation splits across panels
visualise_splits method to visualize the splits created according to the parameters
metrics collection of non-standard scikit-learn performance metrics for evaluation of Performance Ratios and Naive PnL comparisons for the long only and selected macro strategies
machine learning model predictions
signal_optimizer calculation of quantamental predictions based on adaptive hyperparameter and
model selection
PnL - calculation, evaluation and plotting PnLs
make_pnl calculates a daily PnL for a specific signal category
make_long_pnl calculates daily long-only PnL, used for comparison
plot_pnls plot a line chart of cumulative PnL
evaluate_pnls returns a small dataframe of key PnL statistics
signal_heatmap creates a heatmap of signals for a specific PnL across time and sections
plot_pnls plot a line chart of cumulative PnL
create_results_dataframe outputs table with relevant performance statistics for selected signals
JPMaQS Disclaimer
This material has been prepared by Macrosynergy Limited (“Macrosynergy”), a company incorporated in the UK (Company number 06994336), and J.P. Morgan. It is provided for information purposes and
intended for your use only and does not constitute an invitation or offer to subscribe for or purchase any investment product or service. This material has been provided specifically for the use of the recipient
only and must be treated as proprietary and confidential. It may not be passed on, nor reproduced in whole or in part under any circumstances without express written consent from Macrosynergy and J.P.
Morgan. The material provided is not intended to provide a sufficient basis on which to make an investment decision.

The information and opinions contained in this material are for background purposes only and do not purport to be full or complete. No reliance may be placed for any purpose on the information or opinions
contained in this material or their accuracy or completeness. No representation, warranty or undertaking, express or implied, is given as to the accuracy or completeness of the information or opinions
contained in this material by Macrosynergy or J.P. Morgan and no liability is accepted by Macrosynergy or J.P. Morgan for the accuracy or completeness of any such information or opinions. Macrosynergy
and J.P. Morgan believe that the sources of this material are reliable. However, they cannot and do not guarantee, either expressly or implicitly, and accept no liability for, the accuracy, validity, timeliness, mer-
chantability or completeness of any information or data (whether prepared by Macrosynergy, J.P. Morgan or a third party) for any particular purpose or use or that the information or data will be free from error.

Relevant information in respect of Macrosynergy’s compliance with the EU General Data Protection Regulation (Regulation (EU) 2016 679, including our updated privacy notice, is available on our website.

Past performance is not necessarily indicative of future results. This communication may include returns for various indices.

This material is not a research report of J.P. Morgan. J.P. Morgan research analysts may independently use the data services described herein in connection with publishing J.P. Morgan research. The
Macrosynergy asset management advisory services described herein are separate from any business of J.P. Morgan.

J.P. Morgan is a marketing name for investment banking businesses of JPMorgan Chase & Co. and its subsidiaries and affiliates worldwide.