Volatility Risk Premium

Medium · Jun 15, 2026Volatility Risk Premium

A Python research project comparing implied and forward realized volatility, with notebooks and a Streamlit dashboard.

Open Streamlit ↗GitHub ↗

Python
Pandas
NumPy
Streamlit

VIX vs. realized volatility

Historical volatility

Period

Annualized volatility (%)Monthly observation date

VIX
Forward realized volatility

196 monthly observations, 2010-01-29 to 2026-04-22. At the last observation, VIX was 18.92% and forward realized volatility was 9.58%.

View chart data

Annualized volatility (%) by monthly observation date
Monthly observation date	VIX	Forward realized volatility
2010-01-29	24.62	15
2010-02-26	19.5	7.39
2010-03-31	17.59	23.06
2010-04-30	22.05	31.3
2010-05-28	32.07	24.68
2010-06-30	34.54	20.22
2010-07-30	23.5	19
2010-08-31	26.05	14.64
2010-09-30	23.7	11.45
2010-10-29	21.2	14.3
2010-11-30	23.54	8.51
2010-12-31	17.75	10
2011-01-31	19.53	14.43
2011-02-28	18.35	14.47
2011-03-31	17.74	9.74
2011-04-29	14.75	12.48
2011-05-31	15.45	16.21
2011-06-30	16.52	36.74
2011-07-29	25.25	43.48
2011-08-31	31.62	29.85
2011-09-30	42.96	30.05
2011-10-31	29.96	27.64
2011-11-30	27.8	16.35
2011-12-30	23.4	8.69
2012-01-31	19.44	10.55
2012-02-29	18.43	13.14
2012-03-30	15.5	13.54
2012-04-30	17.15	15.99
2012-05-31	24.06	18.44
2012-06-29	17.08	13.08
2012-07-31	18.93	9.98
2012-08-31	17.47	10.67
2012-09-28	15.73	12.77
2012-10-31	18.6	13.31
2012-11-30	15.87	12.24
2012-12-31	18.02	9.93
2013-01-31	14.28	10.97
2013-02-28	15.51	8.44
2013-03-28	12.7	13.19
2013-04-30	13.52	12.08
2013-05-31	16.3	15.14
2013-06-28	16.86	7.51
2013-07-31	13.45	10.27
2013-08-30	17.01	11.39
2013-09-30	16.6	12.29
2013-10-31	13.75	9.13
2013-11-29	13.7	9.92
2013-12-31	13.72	14.41
2014-01-31	18.41	12.34
2014-02-28	14	12.68
2014-03-31	13.88	11.72
2014-04-30	13.41	7.62
2014-05-30	11.4	6.33
2014-06-30	11.57	10.54
2014-07-31	16.95	7.51
2014-08-29	12.09	14.01
2014-09-30	16.31	16.07
2014-10-31	14.03	8.85
2014-11-28	13.33	16.09
2014-12-31	19.2	16.04
2015-01-30	20.97	12.21
2015-02-27	13.34	12.76
2015-03-31	15.29	9.71
2015-04-30	14.55	10.25
2015-05-29	13.84	12.22
2015-06-30	18.23	11.42
2015-07-31	12.12	26.44
2015-08-31	28.43	20.44
2015-09-30	24.5	12.04
2015-10-30	15.07	14.77
2015-11-30	16.13	19.44
2015-12-31	18.21	22.58
2016-01-29	20.2	17.6
2016-02-29	20.55	11.91
2016-03-31	13.95	10.41
2016-04-29	15.7	10.21
2016-05-31	14.19	16.81
2016-06-30	15.63	7.09
2016-07-29	11.87	9.94
2016-08-31	13.42	12.95
2016-09-30	13.29	9.78
2016-10-31	17.06	9.78
2016-11-30	13.33	7.4
2016-12-30	14.04	6.03
2017-01-31	11.99	6.52
2017-02-28	12.92	7.18
2017-03-31	12.37	6.14
2017-04-28	10.82	7.1
2017-05-31	10.41	7.43
2017-06-30	11.18	7.14
2017-07-31	10.26	9.11
2017-08-31	10.59	5.03
2017-09-29	9.51	4.45
2017-10-31	10.18	5.8
2017-11-30	11.28	6.29
2017-12-29	11.04	21.23
2018-01-31	13.54	23.18
2018-02-28	19.85	20.77
2018-03-29	19.97	15.49
2018-04-30	15.93	9.58
2018-05-31	15.43	9.23
2018-06-29	16.09	8.03
2018-07-31	12.83	6.56
2018-08-31	12.86	13.29
2018-09-28	12.12	21.5
2018-10-31	21.23	20.71
2018-11-30	18.07	27.61
2018-12-31	25.42	16.21
2019-01-31	16.57	9.04
2019-02-28	14.78	10.54
2019-03-29	13.71	11.2
2019-04-30	13.12	14.83
2019-05-31	18.71	9.21
2019-06-28	15.08	14.26
2019-07-31	16.12	20.36
2019-08-30	18.98	12.67
2019-09-30	16.24	11.78
2019-10-31	13.22	7.11
2019-11-29	12.62	7.81
2019-12-31	13.78	11.8
2020-01-31	18.84	67.98
2020-02-28	40.11	85.14
2020-03-31	53.54	37.1
2020-04-30	34.15	28.02
2020-05-29	27.51	26.49
2020-06-30	30.43	12.38
2020-07-31	24.46	18.64
2020-08-31	26.41	23.49
2020-09-30	26.37	21.02
2020-10-30	38.02	14.34
2020-11-30	20.57	10.16
2020-12-31	22.75	15.28
2021-01-29	33.09	16.46
2021-02-26	27.95	15.21
2021-03-31	19.4	13.14
2021-04-30	18.61	11.85
2021-05-28	16.76	8.74
2021-06-30	15.83	10.33
2021-07-30	18.24	7.82
2021-08-31	16.48	12.94
2021-09-30	23.14	9.87
2021-10-29	16.26	14.86
2021-11-30	27.19	16.13
2021-12-31	17.22	19.45
2022-01-31	24.83	23.86
2022-02-28	30.15	21.62
2022-03-31	20.56	27.52
2022-04-29	33.4	31.5
2022-05-31	26.19	25.78
2022-06-30	28.71	17.91
2022-07-29	21.33	19.19
2022-08-31	25.87	26.51
2022-09-30	31.62	30.29
2022-10-31	25.88	24.73
2022-11-30	20.58	17.85
2022-12-30	21.67	16.28
2023-01-31	19.4	16.5
2023-02-28	20.7	15.98
2023-03-31	18.7	11.97
2023-04-28	15.78	12.38
2023-05-31	17.94	10.09
2023-06-30	13.59	9.09
2023-07-31	13.63	11.31
2023-08-31	13.57	11.51
2023-09-29	17.52	14.8
2023-10-31	18.14	10.45
2023-11-30	12.92	9.75
2023-12-29	12.45	11.44
2024-01-31	14.35	11.8
2024-02-29	13.4	11.32
2024-03-28	13.01	13.02
2024-04-30	15.65	8.7
2024-05-31	12.92	6.89
2024-06-28	12.44	18.43
2024-07-31	16.36	19.29
2024-08-30	15	12.96
2024-09-30	16.73	12.79
2024-10-31	23.16	10.74
2024-11-29	13.51	14.9
2024-12-31	17.35	12.72
2025-01-31	16.43	17.66
2025-02-28	19.63	41.26
2025-03-31	22.28	42.04
2025-04-30	24.7	14.28
2025-05-30	18.57	9.26
2025-06-30	16.73	9.63
2025-07-31	16.72	10.78
2025-08-29	15.36	11.44
2025-09-30	16.28	13.71
2025-10-31	17.44	12.95
2025-11-28	16.35	8.25
2025-12-31	14.95	12.43
2026-01-30	17.44	13.43
2026-02-27	19.86	17.87
2026-03-31	25.25	11
2026-04-22	18.92	9.58

Monthly observations from the project dataset · January 2010–April 2026.

From the notebook

Full notebook ↗

02_realized_volatility.ipynb · original code cells

Cell 3Realized volatility

import numpy as np

#  ln(P_t / P_{t-1})
merged_df['spx_log_ret'] = np.log(merged_df['spx_close'] / merged_df['spx_close'].shift(1))

# rollingStandardDeviation
merged_df['spx_roll_std'] = merged_df['spx_log_ret'].rolling(window=30).std()

# annualizeDailyVolatility
merged_df['spx_realized_vol'] = merged_df['spx_roll_std'] * np.sqrt(252)

# tail
merged_df[['spx_close', 'spx_log_ret', 'spx_roll_std', 'spx_realized_vol']].tail()

Cell 11Forward alignment

# 1. Shift the realized volatility backwards by 30 trading days
# shiftThirty
merged_df['forward_realized_vol'] = merged_df['spx_realized_vol'].shift(-30)

# 2. dropFinalThirty
master_df = merged_df.dropna(subset=['forward_realized_vol']).copy()


master_dataset = master_df[['spx_close', 'vix_decimal', 'forward_realized_vol']]
master_dataset.to_csv('../src/data/master_dataset.csv')

print("Aligned master dataset exported successfully!")
print(f"Original rows: {len(merged_df)}")
print(f"New rows (should be exactly 30 fewer): {len(master_dataset)}")