Paper: arXiv 2409.19385

Authors: Peilun He, Nino Kordzakhia, Gareth W. Peters, Pavel V. Shevchenko

Abstract

PDSim is an R package that enables users to simulate commodity futures prices using the polynomial diffusion model introduced in Filipovic & Larsson (2016) through both a Shiny web application and R scripts. For user-supplied data, a standalone R routine has been developed to provide joint estimation of state variables and model parameters via the Extended Kalman Filter (EKF) or Unscented Kalman Filter (UKF). With its user-friendly interface, PDSim makes the features of simulations and estimations accessible. To date, it is the only package specifically designed for the simulation and estimation of the polynomial diffusion model. The Schwartz-Smith two-factor model (Schwartz & Smith, 2000) is also available within this package for both simulation and calibration. The package is validated through several tests, including replication of the results in Schwartz & Smith (2000), unit testing of the coverage rate, and verification of the outputs of the main functions.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper contains dense mathematical derivations involving polynomial diffusions, SDEs, and generator theory (Theorem 1), but the software implementation focuses on simulation and estimation tools with limited backtesting on real data, placing it in the ‘Lab Rats’ quadrant.

Research Flowchart

  flowchart TD
  A["Research Goal: Develop R Package for<br>Simulating/Estimating Polynomial Diffusion Models"] --> B["Methodology: Model Implementation<br>Polynomial Diffusion & Schwartz-Smith"]
  B --> C["Methodology: Estimation Technique<br>Extended/Unscented Kalman Filters"]
  C --> D["Methodology: Validation & Testing<br>Replication & Unit Testing"]
  D --> E["Data Input: User-Supplied<br>Commodity Futures Data"]
  D --> F["Computational Process: Shiny App<br>GUI & R Package Backend"]
  E --> F
  F --> G["Key Outcomes: PDSim Package<br>Simulation & Estimation Tools"]
  G --> H["Final Outcome: Validated &<br>User-Friendly R Package"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes the R package PDSim for simulating commodity futures prices and estimating latent factors using the polynomial diffusion model.

Source sentence
In this paper, we introduce the R package PDSim, which is specifically designed to simulate futures prices and estimate latent factors using the polynomial diffusion model.

Assumes 4

  • Arbitrage‑free market and non‑stochastic interest rate market
    Source sentence
    under the assumption of an arbitrage‑free market and non‑stochastic interest rate
  • Both latent factors follow risk‑neutral Ornstein‑Uhlenbeck processes model
    Source sentence
    we assume both χt and ξt follow a risk‑neutral Ornstein‑Uhlenbeck (OU) process
  • Standard errors for futures contracts are evenly spaced data
    Source sentence
    we assume that the standard errors σi, i = 1, . . . , m for futures contracts are evenly spaced
  • High dimensionality of the parameter space makes estimation computationally expensive other
    Source sentence
    Given the high dimensionality of the parameter space, estimation is computationally expensive

Data

  • WTI crude oil futures (weekly) · NFCP R package market-prices public
    2 January 1990 to 14 February 1995
    Source sentence
    we use weekly WTI crude oil futures prices from 2 January 1990 to 14 February 1995, which is the exact same dataset used in [22]. This dataset is available in the R package “NFCP”.
  • WTI crude oil futures (2015–2018) market-prices unknown
    2015 to 2018
    Source sentence
    Table 2 presents another application of PDSim to WTI crude oil futures data for the period 2015–2018.
  • Simulated data (Schwartz‑Smith and polynomial diffusion models) synthetic unknown
    Source sentence
    The core package provides exported functions for simulation as well as for linear and non‑linear filtering.

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The polynomial diffusion model outperforms the Schwartz‑Smith model on WTI data 2015‑2018, achieving a mean RMSE of 0.1921. performance
    Source sentence
    In this case, the polynomial diffusion model outperforms the Schwartz–Smith model, achieving a mean RMSE of 0.1921.
  • Fitting the polynomial diffusion model to WTI data takes approximately 30 minutes using L‑BFGS‑B optimisation. other
    Source sentence
    fitting the WTI crude oil futures data using the polynomial diffusion model takes approximately 30 minutes.
  • The Schwartz‑Smith model provides more accurate estimates than the polynomial diffusion model for the earlier 1990‑1995 WTI dataset. performance
    Source sentence
    the Schwartz–Smith model provides more accurate estimates than the polynomial diffusion model for all contracts.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 5 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

https://github.com/peilun-he/PDSim — the authors' own repository

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (25)

  1. Multi-Factor Polynomial Diffusion Models and Inter-Temporal Futures Dynamics — → 2409.19386
  2. Polynomial diffusions and applications in finance (2016)
  3. NFCP : N-Factor Commodity Pricing Through Term Structure Estimation (2021)
  4. Forecasting, structural time series models and the Kalman filter (1990)
  5. New extension of the Kalman filter to nonlinear systems (1997)
  6. Unscented filtering and nonlinear estimation (2004)
  7. A multifactor polynomial framework for long-term electricity forwards with delivery period (2020)
  8. Short-term variations and long-term dynamics in commodity prices (2000)
and 17 more
  1. The unscented Kalman filter for nonlinear estimation (2000)
  2. Which risk factors drive oil futures price curves? (2020)
  3. Stochastic convenience yield implied from commodity futures and interest rates (2005)
  4. Commodity price forecasts, futures prices, and pricing models (2019)
  5. An N -factor Gaussian model of oil futures prices (2006)
  6. Energy futures prices: term structure models with Kalman filter estimation (2002)
  7. Calibration and filtering for multi factor commodity models with seasonality: incorporating panel data from futures contracts (2013)
  8. Modeling seasonality in agricultural commodity futures (2002)
  9. Stochastic convenience yield and the pricing of oil contingent claims (1990)
  10. A two-factor model for the electricity forward market (2009)
  11. Parameter estimation for a bidimensional partially observed Ornstein–Uhlenbeck process with biological application (2010)
  12. A limited memory algorithm for bound constrained optimization (1995)
  13. The negative pricing of the May 2020 WTI contract (2023)
  14. Jumps in commodity markets (2019)
  15. Jump activity analysis for affine jump-diffusion models: Evidence from the commodity market (2019)
  16. Forecasting commodity prices: GARCH , jumps, and mean reversion (2008)
  17. Polynomial jump-diffusion models (2020)

1 of 25 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.