Datasets:
pretty_name: All Daily Turing Risk Index
language:
- en
license: other
task_categories:
- time-series-forecasting
- tabular-regression
tags:
- finance
- quantitative-trading
- backtesting
- algorithmic-trading
- macroeconomics
- daily
size_categories:
- 10K<n<100K
extra_gated_prompt: >-
This dataset is free to browse and gated for download. Approval is tied to a
Papers With Backtest subscription, which also covers the other datasets in
this organisation and the strategy catalogue at
https://paperswithbacktest.com. Plans and what each one includes:
https://paperswithbacktest.com/pricing
dataset_info:
features:
- name: datetime
dtype: string
- name: turing_risk
dtype: float64
- name: market_risk
dtype: float64
- name: business_risk
dtype: float64
- name: political_risk
dtype: float64
- name: hs
dtype: float64
- name: ma
dtype: float64
- name: ewma
dtype: float64
- name: garch
dtype: float64
- name: ensemble
dtype: float64
- name: vix
dtype: float64
- name: systemic
dtype: float64
- name: turbulence
dtype: float64
- name: recession_6
dtype: float64
- name: recession_12
dtype: float64
- name: recession_24
dtype: float64
- name: cape
dtype: float64
- name: naiim_neg
dtype: float64
- name: aaii_neg
dtype: float64
- name: ads_business_neg
dtype: float64
- name: nonman_outlook_neg
dtype: float64
- name: man_phil_neg
dtype: float64
- name: man_tex_neg
dtype: float64
- name: man_ny_neg
dtype: float64
- name: cfnai_fneg
dtype: float64
- name: zew_sent_neg
dtype: float64
- name: atlanta_unc
dtype: float64
- name: building_index_neg
dtype: float64
- name: consumer_index_neg
dtype: float64
- name: industry_index_neg
dtype: float64
- name: main_index_neg
dtype: float64
- name: retail_index_neg
dtype: float64
- name: services_index_neg
dtype: float64
- name: mics_ics_neg
dtype: float64
- name: mics_icc_neg
dtype: float64
- name: mics_ice_neg
dtype: float64
- name: news_sent_neg
dtype: float64
- name: term_spread
dtype: float64
- name: credit_spread
dtype: float64
- name: corp_bond_distress
dtype: float64
- name: misery_index
dtype: float64
- name: housing_afford_neg
dtype: float64
- name: new_trucks
dtype: float64
- name: new_homes
dtype: float64
- name: cfsec_neg
dtype: float64
- name: us_policy_unc_d
dtype: float64
- name: uk_policy_unc_d
dtype: float64
- name: china_policy_unc_m
dtype: float64
- name: us_market_unc_d
dtype: float64
- name: us_policy_vol_m
dtype: float64
- name: global_policy_unc_m
dtype: float64
- name: us_sovereign_unc_m
dtype: float64
- name: geo_unc_d
dtype: float64
- name: geo_unc_m
dtype: float64
- name: geo_equal_m
dtype: float64
- name: web_search_unc_m
dtype: float64
- name: thinktank_unc_m
dtype: float64
splits:
- name: train
num_examples: 24204
All Daily Turing Risk Index
Daily history of the Turing Risk Index and its market, business, and political risk components.
24,204 rows, 57 columns, covering 1960-01-01 to 2026-04-07. The most recent observation is 2026-04-07.
Why It Matters
This dataset is essential for systematic trading, risk management, and quantitative research focused on market uncertainty:
Macro Overlay: The aggregate index and sub-components (market, business, political risk) provide a consolidated view of systemic uncertainty for market timing and asset allocation.
Recession Modeling: Utilize the multi-horizon recession probabilities (
recession_6,recession_12,recession_24) as leading indicators in economic and investment models.Stress Testing: Combine with portfolio data to stress-test holdings against sudden surges in volatility, credit distress, or geopolitical uncertainty.
Load It
Installation/Upgrade: Ensure you have the latest version of the toolkit:
pip install pwb-toolbox
Access Token Setup: Before running the code, ensure your Hugging Face Access Token is set as an environment variable (HF_ACCESS_TOKEN) or defined in your session, and you are subscribed to the Backtester plan on Papers With Backtest.
from pwb_toolbox import datasets as pwb_ds
df = pwb_ds.load_dataset("All-Daily-TuringRiskIndex")
print(df.iloc[0,:])
Example Output:
datetime 1960-01-01
turing_risk 34.737167
market_risk 37.58584
business_risk 37.796684
political_risk 18.032432
hs 47.265806
ma 45.548835
ewma 45.629371
garch 13.879916
ensemble 43.940312
vix 20.046106
systemic 23.89812
turbulence 10.153223
recession_6 28.024086
recession_12 57.873915
recession_24 63.821508
cape 50.948886
naiim_neg 36.501792
aaii_neg 48.418238
ads_business_neg 30.910628
nonman_outlook_neg 27.353725
man_phil_neg 42.388548
man_tex_neg 38.688323
man_ny_neg 35.776361
cfnai_fneg 52.835876
zew_sent_neg 41.137134
atlanta_unc 27.961265
building_index_neg 39.481653
consumer_index_neg 43.601625
industry_index_neg 32.466125
main_index_neg 30.484709
retail_index_neg 27.853436
services_index_neg 31.143671
mics_ics_neg 35.46754
mics_icc_neg 45.701794
mics_ice_neg 30.523532
news_sent_neg 41.411862
term_spread 57.643777
credit_spread 21.581057
corp_bond_distress 26.381857
misery_index 28.958701
housing_afford_neg 48.310476
new_trucks 51.962154
new_homes 31.686657
cfsec_neg 36.166639
us_policy_unc_d 19.269711
uk_policy_unc_d 19.455555
china_policy_unc_m 42.449101
us_market_unc_d 12.186028
us_policy_vol_m 23.570878
global_policy_unc_m 27.334687
us_sovereign_unc_m 16.975858
geo_unc_d 18.319971
geo_unc_m 19.829667
geo_equal_m 21.486922
web_search_unc_m 29.997329
thinktank_unc_m 25.182874
Columns
This dataset includes data with the following fields:
| Column Name | Description |
|---|---|
| datetime | Observation date (UTC). |
| turing_risk | Headline Turing Risk Index (Aggregate Risk Score). |
| market_risk | Component sub-index for Market Uncertainty. |
| business_risk | Component sub-index for Business Cycle Uncertainty. |
| political_risk | Component sub-index for Political/Geopolitical Uncertainty. |
| hs, ma, ewma, garch, ensemble | Volatility and risk model outputs used in the composite index. |
| vix, systemic, turbulence | Market stress and systemic risk gauges. |
| recession_6, recession_12, recession_24 | Recession probabilities at 6-month, 12-month, and 24-month horizons. |
| cape | Shiller Cyclically Adjusted P/E (CAPE) valuation measure. |
| naiim_neg, aaii_neg, ads_business_neg, nonman_outlook_neg, man_phil_neg, man_tex_neg, man_ny_neg, cfnai_fneg, zew_sent_neg, atlanta_unc | Survey and nowcast sentiment: investment manager and retail investor surveys, the ADS business conditions index, the regional Fed manufacturing and non-manufacturing outlooks, the Chicago Fed national activity index, ZEW sentiment and the Atlanta Fed uncertainty measure. |
| building_index_neg, consumer_index_neg, industry_index_neg, main_index_neg, retail_index_neg, services_index_neg | Business activity diffusion indexes by sector. |
| mics_ics_neg, mics_icc_neg, mics_ice_neg | University of Michigan consumer sentiment variants: the headline index, current conditions and expectations. |
| news_sent_neg | News sentiment component. |
| term_spread, credit_spread, corp_bond_distress, misery_index, housing_afford_neg, new_trucks, new_homes, cfsec_neg | Macro and credit conditions: the yield curve, credit spreads, corporate bond distress, the misery index, housing affordability, truck and home sales and financial conditions. |
| us_policy_unc_d, uk_policy_unc_d, china_policy_unc_m, us_market_unc_d, us_policy_vol_m, global_policy_unc_m, us_sovereign_unc_m, geo_unc_d, geo_unc_m, geo_equal_m, web_search_unc_m, thinktank_unc_m | Policy and geopolitical uncertainty across regions. The \_d suffix marks a daily series and \_m a monthly one. |
Columns ending in \_neg are sign-flipped so that a higher value always means more risk, which is what lets the composite add them together.
Access
Browsing the card and the schema is open to anyone. Downloading the files needs an approved request, tied to a subscription: what each plan includes. The same subscription covers the other datasets in this organisation.
Elsewhere
- Dataset page and coverage charts
- The strategy catalogue, 3,806 papers and 4,837 replicated strategies
pwb-toolbox, the loader used in the snippet aboveawesome-systematic-trading, the replicated strategies with their measured Sharpe- Every dataset in this organisation
Papers With Backtest publishes 32 datasets on the Hub and codes the papers that use them. Every strategy in the catalogue is run over its own full history before it is published, which is where the numbers above come from.