Harvesting the Variance Risk Premium in Nuclear and Energy Equities: A Short-Put Portfolio Derisking Strategy
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Quantitative Finance. The day in quantitative finance: models of markets, risk and prices.
Ruth: I'm Ruth.
Warren: And I'm Warren, here with Iris.
Ruth: We're looking at the paper: "Harvesting the Variance Risk Premium in Nuclear and Energy Equities".
Warren: Here's the gist of the paper. Harvesting the Variance Risk Premium in Nuclear and Energy Equities: A Short-Put Portfolio Derisking Strategy is a study documenting a systematic short-put portfolio strategy designed to harvest the variance risk premium in nuclear and energy equities, which offers income without concentrating raw equity drawdown risk.
Ruth: The strategy demonstrates positive returns and superior risk-adjusted performance compared to a passive stock benchmark over the 2000–2024 period.
Warren: VRP Measurement The variance risk premium is measured by comparing implied volatility to realized volatility using a GARCH(1,1) model.
Ruth: The long-run anchor prevents the GARCH model from collapsing toward zero during quiet periods, which is a flaw of EWMA (omega = 0) that inflates the IV/RV ratio spuriously and worsens threshold-strategy performance.
Title and authors: Ruth: So the paper we're talking about is called "Harvesting the Variance Risk Premium in Nuclear and Energy Equities: A Short-Put Portfolio Derisking Strategy." It’s about using a short put portfolio to manage risk in nuclear and energy stocks.
Warren: Yeah, that title tells you exactly what we’re looking at: harvesting that variance risk premium by using this specific short put portfolio strategy.
Iris: What does it actually mean in plain terms for a listener? It means finding income without taking on the full downside risk of owning the underlying stocks.
Ruth: Right. We're talking about generating cash flow from selling options in exchange for bearing the downside risk below a strike price, which is what this strategy is fundamentally doing.
Warren: It’s about that tradeoff—you get income from selling puts but you take on some defined risk if the underlying asset drops significantly or expires out of the money.
Iris: So, when they talk about "variance risk premium," they are essentially talking about the extra compensation investors demand for that uncertainty in these specific sectors.
Ruth: That’s right. And this paper explores how to systematically capture that premium using a short put portfolio structure to derisk exposure.
Warren: It’s not just a passive buy and hold; it’s an active way to harvest the premium by structuring the trades systematically across a universe of nuclear and energy names.
Iris: So, what kind of systematic approach are we looking at in terms of selection criteria? Are they just picking stocks randomly?
Ruth: No, they use a very defined scoring formula to match each put contract to specific delta and days to expiration targets for every name in the universe.
Warren: That means the system is actively trying to find those specific contracts that fit their desired risk profile before even entering them.
Iris: It sounds like they are filtering based on options characteristics—delta and time remaining—to ensure the trades are within a certain operational range.
Ruth: Precisely, they set eligibility rules for delta between zero point two and four tenths and days to expiration between thirty five and fifty five, with open interest needing to be over ten contracts.
Warren: So you’re not just trading whatever is available; you’re selecting specific options that meet those criteria before entry.
Iris: It sounds like the methodology is more about selecting the right instruments for the job rather than just picking names randomly, which makes sense for an econometrician.
Ruth: Exactly, it's about matching the option characteristics to a target profile for each stock in their universe.
Warren: And this selection process is what allows them to build that systematic short put portfolio, making it repeatable and scalable across their entire universe.
Iris: So before we get into the actual numbers, what’s the big picture here? What's the payoff they are aiming for?
Ruth: The payoff is showing that this strategy works consistently over two thousand to twenty twenty four by delivering positive returns, even in years where the variance risk premium was negative.
Warren: They show that it can deliver eighteen point seven percent annualized return on cash secured capital with a Sharpe ratio of seven point eight under always on conditions.
Iris: That return and risk adjusted performance is what’s really compelling when compared to the stock benchmark, which yields fifteen point four percent with a Sharpe ratio of zero point eight one.
Ruth: That difference in risk adjusted performance is huge, by a factor of ten, suggesting that this method for harvesting the premium is very effective.
Warren: It implies that nuclear focused equity options carry a positive estimated risk premium that income seeking investors can harvest systematically, offering stable uncorrelated cash flows to diversify their portfolio.
Iris: So the payoff is showing a systematic way to generate income without taking on all the raw equity risk, which is a key finding for anyone looking at long term energy exposure.
Ruth: And that’s what we need to hear about this paper on harvesting the variance risk premium in nuclear and energy equities.
Warren: And that's what it really boils down to.
The paper's summary: Ruth: Now let’s look deeper into the summary of the paper, because it explains how they are actually putting all these pieces together into a working strategy.
Warren: It lays out the mechanism for building this systematic short put portfolio, which is essentially matching options to specific risk targets for every name in their universe.
Iris: The summary explains that the unconditional strategy involves selling the best fit qualifying put for every name with a tradable contract, which is what makes it so flexible.
Ruth: That flexibility is what lets them use this unconditional approach, but they also talked about how threshold strategies can reduce trade count substantially but suffer higher portfolio volatility because fewer names qualify each month.
Warren: So the summary explains the trade off between a high-volume strategy and a lower volume one.
Iris: That’s right. And it highlights that the unconditional approach achieves an annual portfolio return of eighteen point seven percent with a Sharpe ratio of seven point eight under always on conditions, while threshold strategies achieve similar total returns but suffer higher portfolio volatility due to reduced diversification.
Ruth: So the summary clearly lays out that for this specific study, the unconditional strategy is the one they focus on because it’s much better for risk management.
Warren: And that unconditional strategy has zero maximum drawdown within the sample period, meaning no calendar month produced a negative portfolio return in their sample.
Iris: That zero maximum drawdown is a strong statement about its resilience during market stress, which is what we want to hear when you’re trying to manage risk.
Ruth: It means that even though there were negative variance risk premium years, the positions entered in earlier high variance periods still ended up expiring profitably through those episodes.
Warren: So the summary explains that every calendar year was profitable for the put strategy in their sample period.
Iris: And it’s a very strong statement about its consistency across different market cycles, which is what makes this result feel robust.
Ruth: So to recap, the summary shows that there’s a consistent positive return even when things look rough from a risk perspective.
Warren: That's the core finding of the paper on harvesting the variance risk premium in nuclear and energy equities.
Iris: And it really lays out how this strategy works on a day to day basis for someone trying to understand what’s actually happening.
The paper's improvements: Ruth: Let’s shift gears now and look at the suggested improvements the authors suggest for strengthening this strategy, because they always have some ideas on how to make it better.
Warren: They propose moving towards a GARCH filtered conditional variant by using that implied volatility over realized volatility being greater than one point one zero as a condition to achieve a Sharpe ratio of two point six three.
Iris: That filtering seems like a smart way to adjust the strategy based on market conditions, allowing the system to adjust based on whether the implied volatility is higher or lower or higher or lower.
Ruth: It suggests that you can enter only when that ratio crosses that one point one zero threshold.
Warren: So you’re entering only when the market is showing a condition where implied volatility is significantly greater than realized volatility, which signals a positive regime.
Iris: That filtering helps identify those specific market conditions where the premium is actually present and allows for more precise entry timing based on that signal.
Ruth: It helps pinpoint when to go long rather than just entering whenever the standard rules are met.
Warren: It refines the entry timing based on a specific volatility regime condition, which should help manage risk better during volatile periods.
Iris: That precision in timing is what an econometrician cares about; it’s about making sure that the variation you are measuring is actually driving the outcome.
Ruth: It moves from just entering whenever the standard rules are met to entering only when that specific volatility condition is met, which should improve performance metrics.
Warren: So this adjustment based on a specific volatility regime condition should help manage risk better during volatile periods.
Conclusion: Ruth: So, to wrap up everything we’ve covered with the paper "Harvesting the Variance Risk Premium in Nuclear and Energy Equities: A Short-Put Portfolio Derisking Strategy." We’ve seen how this strategy works from the initial setup to its performance metrics.
Warren: In short, it shows that there is a systematic way to generate income without taking on all the raw equity risk that comes with owning them directly.
Iris: The main implication is that nuclear focused equity options carry a positive estimated risk premium that income seeking investors can harvest systematically.
Pennsylvania State University
q-fin.PM
Submitted: 2026-09-01
Updated: 2026-11-13
Comments: 4 pages, 5 figures, 1 table. Accepted summary, 2026 ANS Winter Conference and Expo (American Nuclear Society), November 2026
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
Importance score: 72/100
The gist: Harvesting the Variance Risk Premium in Nuclear and Energy Equities: A Short-Put Portfolio Derisking Strategy is a study documenting a systematic short-put portfolio strategy designed to harvest the…
Key concepts
- Variance Risk Premium (VRP)
- This is the difference between the implied volatility of options and the actual realized volatility of the underlying assets. The paper measures this by comparing implied volatility to realized volatility using a GARCH(1,1) model to identify periods where options are mispriced relative to market movement.
- Short-Put Portfolio Strategy
- This involves systematically selling put options on a basket of nuclear and energy stocks. The goal is not just income but risk management; by selling puts, the investor collects premium while accepting defined downside risk below the strike price, aiming to profit from the expected volatility.
- GARCH(1,1) Model
- This is a statistical model used to estimate volatility by looking at past volatility and past squared residuals. The paper uses it to calculate realized volatility from historical option prices, which is crucial for accurately measuring the variance risk premium in nuclear and energy equities.
- Threshold Strategy
- This is a method where trades are only executed if certain market conditions are met, such as when implied volatility exceeds a certain multiple of realized volatility. This approach reduces the number of trades but can increase portfolio volatility compared to an unconditional approach.
Terminology
Summary
Harvesting the Variance Risk Premium in Nuclear and Energy Equities: A Short-Put Portfolio Derisking Strategy is a study documenting a systematic short-put portfolio strategy designed to harvest the variance risk premium in nuclear and energy equities, which offers income without concentrating raw equity drawdown risk. p. 0 Harvesting the Variance Risk Premium in Nuclear and Energy Equities:
A Short-Put Portfolio Derisking Strategy
Jilang Miao∗
, Nonna Sorokina†
∗Ken and Mary Alice Lindquist Department of Nuclear Engineering, Pennsylvania State University, University Park, PA 16802
†Division of Business, Pennsylvania State University, Dunmore, PA 18512
INTRODUCTION
Nuclear energy plays a growing role in long-term decarbonization plans, yet private capital flows into nuclear projects
remain constrained by the sector’s distinctive risk profile: high
up-front capital requirements, decade-scale construction timelines, and residual regulatory and political uncertainty. Firms
in controversial industries face higher costs of equity [1], and
nearly all nuclear projects in Europe rely on government support due to limited private capital participation [2]. CONCLUSIONS
We document a positive and economically large variance
risk premium in nuclear and energy equity options and show
that a simple equal-weight short-put portfolio captures it consistently in this sample over 2000–2024. The strategy achieves
19% gross annual return on cash-secured capital with no negative monthly returns in sample, outperforming a long-only
stock benchmark on a risk-adjusted basis by a factor of ten on
the Sharpe ratio before transaction costs and the risk-free rate.Where the paper says it
Where the paper says it
VRP Measurement
The variance risk premium is measured by comparing implied volatility to realized volatility using a GARCH(1,1) model. The long-run anchor prevents the GARCH model from collapsing toward zero during quiet periods, which is a flaw of EWMA (omega = 0) that inflates the IV/RV ratio spuriously and worsens threshold-strategy performance. p. 0 90, and ωi calibrated per stock to match
each ticker’s historical variance: ωi = r
2
i(1−α−β). This longrun anchor prevents GARCH from collapsing toward zero
during quiet periods, a flaw of EWMA (ω = 0) that inflates
the IV/RV ratio spuriously and worsens threshold-strategy
performance. 60] and DTE ∈ [15, 45]. p. 0 We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45]. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45]. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45].Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Strategy Construction
The strategy involves constructing a systematic short-put portfolio for a universe of approximately 45 nuclear and energy-adjacent equities. The selection process requires matching a target of 0. p. 0 The VRP on
date t is IVATM − σˆ
realized
t+h where σˆ
realized
t+h
is the forward 21-day
realized vol. Put Selection
On each entry date, we select the single put contract per
security that best matches a target of 0. Put Selection
On each entry date, we select the single put contract per
security that best matches a target of 0. 30-delta and 45-day
DTE, scored as:
score =
|δ − 0. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45]. Red bars: average ratio ≥ 1. with eligibility restricted to DTE ∈ [35, 55], |δ| ∈ [0. with eligibility restricted to DTE ∈ [35, 55], |δ| ∈ [0. 40],
and open interest > 10 contracts. 40],
and open interest > 10 contracts. Exit Rules
Positions are closed at whichever of three conditions occurs first: (i) premium decays to 50% of entry mid-price (takeprofit), (ii) mid-price reaches twice the entry premium (stoploss), or (iii) DTE ≤ 15 (early expiration roll).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the gains come from
The unconditional strategy, which sells the best-fit qualifying put for every name with a tradable contract, achieves an annual portfolio return of 18.7% with a Sharpe ratio of 7. p. 0 We document (i) statistically significant positive VRP in the majority of names with
option coverage, (ii) portfolio returns of 18. 7% annually with
a Sharpe ratio of 7. We document (i) statistically significant positive VRP in the majority of names with
option coverage, (ii) portfolio returns of 18. 7% annually with
a Sharpe ratio of 7. We document (i) statistically significant positive VRP in the majority of names with
option coverage, (ii) portfolio returns of 18. 7% annually with
a Sharpe ratio of 7. 7% annually with
a Sharpe ratio of 7. 8 under an unconditional (always-on) strategy, versus 15. Put strategies positive in all 25 years; stock falls −37%
in 2008. The unconditional strategy’s high Sharpe reflects two
complementary features: (i) the positive average VRP in nuclear and energy options, and (ii) strong diversification across
64,514 individual trades, compressing portfolio return variance relative to any single position. CAP-10 and Universe Concentration
CAP-10 and unconditional EW produce nearly identical
results because the always-on strategy activates 30–43 tickers
per month; equal weight is already below 10% so the cap
never binds. The threshold variant does reduce active names
substantially (typically 5–15 per month given the IV/RV filter),
which is precisely why it has higher portfolio volatility (4. CAP-10 and Universe Concentration
CAP-10 and unconditional EW produce nearly identical
results because the always-on strategy activates 30–43 tickers
per month; equal weight is already below 10% so the cap
never binds. The threshold variant does reduce active names
substantially (typically 5–15 per month given the IV/RV filter),
which is precisely why it has higher portfolio volatility (4. CAP-10 and Universe Concentration
CAP-10 and unconditional EW produce nearly identical
results because the always-on strategy activates 30–43 tickers
per month; equal weight is already below 10% so the cap
never binds. The threshold variant does reduce active names
substantially (typically 5–15 per month given the IV/RV filter),
which is precisely why it has higher portfolio volatility (4.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Performance Results
The put portfolio delivers 18.7% annualized return on cash-secured capital with a 2. p. 1 Trade return = pnl/ strike (cash-secured; pnl = entry
mid − exit mid in $/share). The option and stock portfolios
deploy equal notional capital per name; the comparison is on
a cash-secured basis throughout. 7% annualized return at
2. 4% annualized volatility (Sharpe 7. The VRP is positive in 19 of 25 calendar years (Figure 3). The six negative years — 2000, 2002, 2008, 2011, 2018, and
2024 — all correspond to periods of elevated realized volatility
that temporarily exceeded implied levels; 2008 shows the
0.Where the paper says it
Where the paper says it
Where the paper says it
Limitations and Extensions
Several limitations include using option mid-prices throughout, which may understate gap risk on adverse overnight moves. The model assumes cash-secured capital earns no T-bill return, understating opportunity cost particularly post-2022. p. 3 The backtest uses option mid-prices throughout; realistic bid-ask execution costs for short puts, which can be substantial for less liquid
names, would reduce reported returns. The cash-secured capital earns no T-bill return in our model, understating opportunity
cost particularly post-2022; incorporating the risk-free rate
would lower Sharpe ratios. The cash-secured capital earns no T-bill return in our model, understating opportunity
cost particularly post-2022; incorporating the risk-free rate
would lower Sharpe ratios. The universe is fixed at today’s list
of survivors, introducing survivorship and look-ahead bias; a
point-in-time reconstruction is needed before the performance
claims can be treated as investable. Exit prices assume fills at
the first daily mid satisfying the stop or take-profit condition,
which may understate gap risk on adverse overnight moves. On the model side, GARCH parameters are fixed at consensus
values rather than estimated per ticker, and the many hardcoded design choices (exit thresholds, target delta and DTE,
stop-loss multiples) have not been jointly optimized; a systematic design-optimization framework could identify whether
the current parameterization is robust or leaves performance
on the table.Where the paper says it
Where the paper says it
Where the paper says it
Win Rates and Crisis Performance
Across all 45 tickers and 64,514 always-on trades, the overall win rate is 88.3%, with annual win rates ranging from 84. p. 3 Win Rates and Crisis Performance
Across all 45 tickers and 64,514 always-on trades, the
overall win rate is 88. Annual win rates range from 84. Win Rates and Crisis Performance
Across all 45 tickers and 64,514 always-on trades, the
overall win rate is 88. Annual win rates range from 84. The VRP is positive in 19 of 25 calendar years (Figure 3). The six negative years — 2000, 2002, 2008, 2011, 2018, and
2024 — all correspond to periods of elevated realized volatility
that temporarily exceeded implied levels; 2008 shows the
0. The VRP is positive in 19 of 25 calendar years (Figure 3). The six negative years — 2000, 2002, 2008, 2011, 2018, and
2024 — all correspond to periods of elevated realized volatility
that temporarily exceeded implied levels; 2008 shows the
0. 8% (2002),
because positions entered in earlier high-VRP periods continued to expire profitably through those episodes. CONCLUSIONS
We document a positive and economically large variance
risk premium in nuclear and energy equity options and show
that a simple equal-weight short-put portfolio captures it consistently in this sample over 2000–2024.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Conclusions
The results document a positive and economically large variance risk premium in nuclear and energy equity options, showing that a simple equal-weight short-put portfolio captures it consistently over 2000–2024. These findings suggest that nuclear-focused equity options carry a positive estimated risk premium that income-seeking investors can harvest systematically, providing a complement to direct nuclear equity ownership. p. 3 Every one of the 25 calendar years was profitable. These results suggest that nuclear-focused equity options
carry a positive estimated risk premium that income-seeking
investors can harvest systematically, providing a complement
to direct nuclear equity ownership. These results suggest that nuclear-focused equity options
carry a positive estimated risk premium that income-seeking
investors can harvest systematically, providing a complement
to direct nuclear equity ownership. By offering stable, uncorrelated cash flows, the strategy may broaden the investor base
for nuclear-adjacent equities, a step toward closing the private
financing gap in nuclear energy development.Where the paper says it
Where the paper says it
Acknowledgments
The authors thank the Pennsylvania State University Institute for Computational and Data Sciences (ICDS) for computational resources support. N. Sorokina acknowledges support from the U. p. 3 ACKNOWLEDGMENTS
The authors thank the Pennsylvania State University Institute for Computational and Data Sciences (ICDS) for computational resources support. Sorokina acknowledges support
from the U. Sorokina acknowledges support
from the U. Department of Energy, Office of Nuclear Energy,
under Award No. Department of Energy, Office of Nuclear Energy,
under Award No. DE-NE0009381, and from ICDS (RRID:
SCR_025154).Where the paper says it
Where the paper says it
Where the paper says it
References
The references cited in the paper are listed below. The list includes sources such as Fernandez, Stein and Lo [3], Bakshi and Kapadia [4], Carr and Wu [5], Bollerslev, Tauchen and Zhou [6], Generalized Autoregressive Conditional Heteroskedasticity [7], and Lee et al. p. 3 ZHOU, “Expected Stock Returns and Variance Risk Premia,” Review
of Financial Studies, 22, 11, 4463–4492 (2009). BOLLERSLEV, “Generalized Autoregressive Conditional Heteroskedasticity,” Journal of Econometrics, 31, 3,
307–327 (1986). 01183v1 [q-fin. PM] 1 Sep 2026 Financial engineering offers complementary tools for
income-seeking institutional investors who want nuclear sector exposure without concentrating in raw equity drawdown
risk. Fernandez, Stein and Lo [3] showed that a mega-fund
structure can diversify the idiosyncratic risk of early-stage
biotech portfolios, reducing investor uncertainty enough to
crowd in private capital.Where the paper says it
Where the paper says it
Where the paper says it
Data and Universe
The data sources used include OptionMetrics IvyDB for daily option data and the CRSP Daily Stock File for daily stock data. The universe consists of 45 tickers drawn from firms identified as nuclear plant operators via NRC plant-ownership records and constituents of nuclear-focused ETFs. p. 0 Daily stock data come from the
CRSP Daily Stock File: prices, returns, and trading volume
identified by PERMNO. Our universe consists of 45 tickers drawn from two independently constructed datasets: firms identified as nuclear
plant operators via NRC plant-ownership records (utilities
with material nuclear generation shares), and constituents of
nuclear-focused ETFs (uranium miners, SMR developers, nuclear service contractors). Our universe consists of 45 tickers drawn from two independently constructed datasets: firms identified as nuclear
plant operators via NRC plant-ownership records (utilities
with material nuclear generation shares), and constituents of
nuclear-focused ETFs (uranium miners, SMR developers, nuclear service contractors). Representative names include CEG,
NEE, CCJ, UEC, NXE, UUUU, BWXT, NRG, and SMR. Coverage spans 2000–2024. Six of the 45
tickers lack sufficient option coverage and are excluded; 39
tickers enter the signal construction and backtest. Coverage spans 2000–2024. Six of the 45
tickers lack sufficient option coverage and are excluded; 39
tickers enter the signal construction and backtest.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
The VRP measurement estimates realized volatility using GARCH(1,1) [7]. The comparison is made against an ATM implied volatility proxy derived from the open-interest-weighted average IV of puts with δ ∈ [0. p. 0 Parameters α and β are fixed at consensus values
rather than estimated by MLE per ticker; per-stock estimation
is left for future work. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. Parameters α and β are fixed at consensus values
rather than estimated by MLE per ticker; per-stock estimation
is left for future work. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. Parameters α and β are fixed at consensus values
rather than estimated by MLE per ticker; per-stock estimation
is left for future work. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. Put Selection
On each entry date, we select the single put contract per
security that best matches a target of 0. 30-delta and 45-day
DTE, scored as:
score =
|δ − 0. Put Selection
On each entry date, we select the single put contract per
security that best matches a target of 0. 30-delta and 45-day
DTE, scored as:
score =
|δ − 0. Put Selection
On each entry date, we select the single put contract per
security that best matches a target of 0. 30-delta and 45-day
DTE, scored as:
score =
|δ − 0. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45]. We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45].Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Exit Rules
Positions are closed at whichever of three conditions occurs first: premium decays to 50% of entry mid-price (takeprofit), mid-price reaches twice the entry premium (stoploss), or DTE ≤ 15 for early expiration roll. All positions are cash-secured, with notional capital per share equaling the put strike. p. 1 Exit Rules
Positions are closed at whichever of three conditions occurs first: (i) premium decays to 50% of entry mid-price (takeprofit), (ii) mid-price reaches twice the entry premium (stoploss), or (iii) DTE ≤ 15 (early expiration roll). All positions are
cash-secured; notional capital per share equals the put strike. Exit Rules
Positions are closed at whichever of three conditions occurs first: (i) premium decays to 50% of entry mid-price (takeprofit), (ii) mid-price reaches twice the entry premium (stoploss), or (iii) DTE ≤ 15 (early expiration roll). All positions are
cash-secured; notional capital per share equals the put strike. All positions are
cash-secured; notional capital per share equals the put strike. Early assignment is not modeled; since puts are selected at
0. All positions are
cash-secured; notional capital per share equals the put strike. Early assignment is not modeled; since puts are selected at
0. Early assignment is not modeled; since puts are selected at
0. 30δ (out-of-the-money), early exercise is rarely optimal for
the holder, though this remains a modeling simplification.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Portfolio Performance
The unconditional strategy's performance shows an annual return of 18.7% with a Sharpe ratio of 7.8, while the stock benchmark yields 15. p. 2 2 loses money in 2011 (−1. 2%) while 1. 2 loses money in 2011 (−1. 2%) while 1. CONCLUSIONS
We document a positive and economically large variance
risk premium in nuclear and energy equity options and show
that a simple equal-weight short-put portfolio captures it consistently in this sample over 2000–2024. The strategy achieves
19% gross annual return on cash-secured capital with no negative monthly returns in sample, outperforming a long-only
stock benchmark on a risk-adjusted basis by a factor of ten on
the Sharpe ratio before transaction costs and the risk-free rate. CONCLUSIONS
We document a positive and economically large variance
risk premium in nuclear and energy equity options and show
that a simple equal-weight short-put portfolio captures it consistently in this sample over 2000–2024. The strategy achieves
19% gross annual return on cash-secured capital with no negative monthly returns in sample, outperforming a long-only
stock benchmark on a risk-adjusted basis by a factor of ten on
the Sharpe ratio before transaction costs and the risk-free rate. 81) for the stock benchmark. The strategies achieve similar total returns but the put portfolio’s risk is approximately eight times lower, with a 47%
maximum drawdown for equities versus zero for the put strategy. 81) for the stock benchmark. The strategies achieve similar total returns but the put portfolio’s risk is approximately eight times lower, with a 47%
maximum drawdown for equities versus zero for the put strategy. 90, and ωi calibrated per stock to match
each ticker’s historical variance: ωi = r
2
i(1−α−β). This longrun anchor prevents GARCH from collapsing toward zero
during quiet periods, a flaw of EWMA (ω = 0) that inflates
the IV/RV ratio spuriously and worsens threshold-strategy
performance. 7% annually with
a Sharpe ratio of 7. 8 under an unconditional (always-on) strategy, versus 15. 63, MaxDD 7. 7%) is
materially better than 1. 7% annually with
a Sharpe ratio of 7. 8 under an unconditional (always-on) strategy, versus 15. The threshold strategy
($12) also outperforms stock on a risk-adjusted basis with a
47% lower maximum drawdown. Note that the stock curve’s
47% peak-to-trough drawdown is measured on monthly returns
and accumulates across the 2007–2009 decline; the annual bar
chart (Figure 4) reflects individual calendar-year returns, with
2008 showing −37%.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
The strategy achieves 19% gross annual return on cash-secured capital with no negative monthly returns in the sample, outperforming a long-only stock benchmark on a risk-adjusted basis by a factor of ten. These results suggest that nuclear-focused equity options carry a positive estimated risk premium that income-seeking investors can harvest systematically. p. 3 Every one of the 25 calendar years was profitable. These results suggest that nuclear-focused equity options
carry a positive estimated risk premium that income-seeking
investors can harvest systematically, providing a complement
to direct nuclear equity ownership. These results suggest that nuclear-focused equity options
carry a positive estimated risk premium that income-seeking
investors can harvest systematically, providing a complement
to direct nuclear equity ownership. By offering stable, uncorrelated cash flows, the strategy may broaden the investor base
for nuclear-adjacent equities, a step toward closing the private
financing gap in nuclear energy development. The strategies achieve similar total returns but the put portfolio’s risk is approximately eight times lower, with a 47%
maximum drawdown for equities versus zero for the put strategy. Every calendar year was profitable for the put strategy,
including 2002 (+31%), 2008 (+26%), 2011 (+20%), and
2018 (+12%), all years in which the stock benchmark lost
value (Figure 4).Where the paper says it
Where the paper says it
Where the paper says it
Pruning redundant reasoning tokens
The unconditional strategy's high Sharpe reflects two complementary features: the positive average VRP in nuclear and energy options and strong diversification across 64,514 individual trades. The threshold strategies reduce trade count substantially but suffer higher portfolio volatility because fewer names qualify each month. The unconditional strategy's zero maximum drawdown means no calendar month produced a negative portfolio return in the sample.
How the training signal is built
The VRP measurement estimates realized volatility using GARCH(1,1) [7]. The comparison is made against an ATM implied volatility proxy derived from the open-interest-weighted average IV of puts with δ ∈ [0.40, 0.60] and DTE ∈ [15, 45]. Put selection is based on a score defined by δ − 0.30 / 0.1 + DTE − 45 / 10.
The unconditional strategy's high Sharpe reflects two complementary features: the positive average VRP in nuclear and energy options and strong diversification across 64,514 individual trades. The threshold strategies reduce trade count substantially but suffer higher portfolio volatility because fewer names qualify each month.
The unconditional strategy's performance shows an annual return of 18.7% with a Sharpe ratio of 7.8, while the stock benchmark yields 15.4% return and a Sharpe ratio of 0.81. The threshold strategies achieve similar total returns but suffer higher portfolio volatility due to reduced diversification.
Across all 45 tickers and 64,514 always-on trades, the overall win rate is 88.3%, with annual win rates ranging from 84.6% (2008) to 92.7% (2004). All six negative-VRP years nonetheless generated positive put portfolio returns, ranging from 12. p. 3 7% (2004), showing no secular degradation over
the 25-year sample. All six negative-VRP years (2000, 2002,
2008, 2011, 2018, 2024) nonetheless generated positive put
portfolio returns, ranging from 12. The VRP is positive in 19 of 25 calendar years (Figure 3). The six negative years — 2000, 2002, 2008, 2011, 2018, and
2024 — all correspond to periods of elevated realized volatility
that temporarily exceeded implied levels; 2008 shows the
0.Where the paper says it
Where the paper says it
The VRP measurement estimates realized volatility using GARCH(1,1) [7]. The comparison is made against an ATM implied volatility proxy derived from the open-interest-weighted average IV of puts with δ ∈ [0.40, 0.60] and DTE ∈ [15, 45]. Put selection is based on a score defined by δ − 0.3 / 1 + DTE − 45 / 10. p. 0 We compare σˆ t
to an ATM implied
volatility proxy: the open-interest-weighted average IV of
puts with |δ| ∈ [0. 60] and DTE ∈ [15, 45].Where the paper says it
Positions are closed at whichever of three conditions occurs first: premium decays to 50% of entry mid-price (takeprofit), mid-price reaches twice the entry premium (stoploss), or DTE ≤ 15 for early expiration roll. All positions are cash-secured, with notional capital per share equaling the put strike. The early assignment is not modeled as puts are selected at 0.3delta (out-of-the-money). p. 1 Early assignment is not modeled; since puts are selected at
0. 30δ (out-of-the-money), early exercise is rarely optimal for
the holder, though this remains a modeling simplification.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Systematic Risk Harvesting Capability: The improved AI system can systematically harvest positive variance risk premium by constructing a systematic short-put portfolio for nuclear and energy equities, as documented in the paper. This allows the system to generate
recurring income
by selling options in exchange for bearing downside risk below a strike price. -
Robust Portfolio Strategy Execution: The system can execute various portfolio strategies including an unconditional equal-weight strategy, a capped version (CAP-10/CAP-20), and threshold strategies based on ATM Implied Volatility divided by GARCH Realized Volatility. This provides flexibility in managing exposure based on market conditions, such as entering only when
ATM IV / GARCH RV ≥ 1.10.
-
Risk Mitigation and Drawdown Control: The AI can implement risk management rules that result in a
zero
maximum drawdown for the unconditional strategy within the sample period. This is achieved by setting hard caps on individual trade losses at a stoploss and exiting positions whenpremium decays to 50% of entry mid-price
or when themid-price reaches twice the entry premium.
-
Signal Identification and Selection: The system can identify statistically significant positive variance risk premiums across a universe of equities by calculating t-statistics for mean IV minus forward RV on GARCH signal days. This allows the AI to select specific nuclear and energy names, such as
NRG (t = 12.7), NXE (t = 12.1), UUUU (t = 10.0),
for targeted option coverage. -
Dynamic Universe Adaptation: The system can be designed to adapt its entry criteria based on the realized volatility environment, specifically by using a
GARCH-filtered conditional variant (IV/RV ≥ 1.10)
to achieve a Sharpe ratio of 2.6, demonstrating an ability to adjust strategy based on market regime changes.
Abstract
We study whether nuclear and energy-adjacent equity options carry a harvestable variance risk premium. Using CRSP and OptionMetrics data for 2000-2024 on a universe of 49 optioned securities under 41 ticker labels, at-the-money implied volatility exceeds the volatility realized over the following 21 trading days in 18 of 25 years, and 29 of 41 tickers show a positive premium on signal days with t-statistics above 2. A fully mechanical cash-secured short-put book sells one put per name per month near 0.25 delta and 45 days to expiry, and exits on the first touch of take-profit at half premium, stop-loss at twice premium, or a 0.50-delta stop, marked daily at validated quote midpoints. Equal-weight and inverse-volatility weightings earn 8.3% and 7.1% a year at Sharpe ratios of 0.98 and 0.99 with maximum drawdowns of 16% and 13%, versus 15.5% a year, a Sharpe ratio of 0.80, and a 46% drawdown for an equal-weight stock book; in 2008 the put book loses 9.5% against the stock book's 37.2%, and the monthly correlation between the two is 0.62. Performance is reported before transaction costs, on a fixed present-day universe.
Related papers
- Bayesian Parametric Portfolio Policies
- Mean--Variance Portfolio Selection by Continuous-Time Reinforcement Learning: Algorithms, Regret Analysis, and Empirical Study
- Beyond Co-Movement: Locality by Exposures Enables a Joint Factor-Graph Framework for Portfolio Diversification
- Interpretable Allocation and Model-Predictive Rebalancing of Equity Portfolios, Tested on Real and Synthetic Markets
- Penalize the Common Trade: A Collision-Aware Layer for Learned Portfolios
- Mitigating Return Drag and Regime Whipsaws in Sector Rotation: An Asymmetric Downside-Aware Nested Clustered Optimization Framework